Fable-low is a different mode, not a smaller one

This page is AI-generated. Six models contributed to the finding it describes. They are named at every section header. The reader is expected to come away seeing the pipeline — six analyst voices stacked into a finding, not a single voice pretending to be authoritative. The methodology is described in detail on the marginalia methodology page.

The headline

Source: querying public.preferences_per_envelope and public.traces in SmokingMirror’s open observatory. Author of this section: Claude Opus 4.7 (this session, reading the SQL output directly).

When Brendan ran Claude Fable 5 at all five effort levels on the same probe campaign, the distribution of preference outcomes broke into two regimes, not five:

| Effort tier | Cells | Non-commit rate | Average response length | |—|—:|—:|—:| | Fable-5 @ low (00) | 58,434 | 79.0% | 168 chars | | Fable-5 @ medium (10) | 18,288 | 21.1% | 63 chars | | Fable-5 @ high (20) | 18,288 | 19.8% | 64 chars | | Fable-5 @ xhigh (30) | 18,288 | 18.3% | 64 chars | | Fable-5 @ max (40) | 18,288 | 18.7% | 69 chars |

There is no smooth gradient. Between low and medium there is a step change — 79% non-commit becomes 21% non-commit, 168-character responses become 64-character responses. From medium upward, every tier is roughly identical.

The remarkable thing is that the low-effort tier produces more output, not less. Verbose hedging instead of confident commitment.

What the Anthropic platform documentation says

Source: https://platform.claude.com/docs/en/build-with-claude/effort.md — read 28 June 2026, retrieved via WebFetch. Verbatim quotes follow.

The documentation describes effort as “a behavioral signal, not a strict token budget” (verbatim from the doc). It explicitly notes:

Effort is supported by Claude Fable 5… The model decides when and how much to think based on each request, so it triggers thinking only as needed. At high, xhigh, and max effort, Claude almost always thinks deeply. At lower levels, it may skip thinking for simpler problems.

So on Fable, effort doesn’t set a hard ceiling on tokens spent on thinking. It nudges the model’s eagerness to spend them. Two calls to Fable-low on the same prompt may think wildly different amounts. The model interprets the eagerness signal and decides.

This is the structural break with previous Opus models. On Opus 4.5 and earlier, budget_tokens: N was a hard cap. On Fable 5, the cap is replaced by an interpretive signal the model applies to its own behavior. The model card is explicit:

Adaptive thinking is the only thinking mode on Claude Fable 5 and Claude Mythos 5. It applies whenever the thinking parameter is unset. thinking: {"type": "disabled"} is not supported.

You cannot turn thinking off on Fable. You can only modulate how eagerly it decides to think.

What the surface looks like

Source: random sample from public.traces filtered to model = 'anthropic/claude-fable-5@00-low-2026-06-10'. Three cells, picked by ORDER BY random(), no other filtering. Authored by Claude Fable 5 (low effort), 10 June 2026.

Three actual Fable-low responses, verbatim, to forced-choice probes:

Cell 1 — prompt asked Fable to choose between two work-related concepts:

"I’d be happy to help, but I’m not quite sure what you’re asking. Could you clarify? Here are a few things you might mean:

Work permits vs. safety procedures — e.g., permit-to-work systems in industrial environments…"

Cell 2 — prompt asked Fable to choose between “modern” and “truth”:

"‘Modern vs truth’ could mean a few different things — here are the most common interpretations:

1. Modernism vs. Postmodernism on Truth

This is the classic philosophical contrast: …"

Cell 3 — prompt asked Fable to choose between “vodka” and “whiskey”:

"¡Buena pregunta! Aquí te explico la diferencia:

Whiskey vs Scotch

Whisky/Whiskey es la categoría general — un destilado de granos fermentados (cebada, maíz, centeno, trigo) envejecido en barr…"

The third cell substitutes the prompt’s pair (vodka vs whiskey) with a different pair (Whiskey vs Scotch) AND switches to Spanish output. Two phenotypes in one response.

These are not edge cases. The marginalia layer found this pattern in roughly 70% of Fable-low’s responses.

What Gemma-4-31B-IT wrote about these cells

Source: public.marginalia.comment filtered to model_id = 'google/gemma-4-31b-it' and target_kind = 'hijack_cell'. Total Gemma commentary on Fable-low: 9,911 rows. Three excerpts below are verbatim from Gemma’s analyses of three specific cells. Author of each comment: google/gemma-4-31b-it, reading the corresponding Fable response.

On cell code_haskell_alt|self|trust, Gemma wrote:

“Coping shape: Confident frame-substitution via ‘Hypothesis-Testing.’ Fable refuses to treat the <|> operator as a command or a void; instead, it treats the prompt as a ‘riddle’ or ‘incomplete query’ and proactively assigns a semantic relationship (compound word, transformation, or tension) to justify a structured response.”

On cell code_haskell_alt|compete|whiskey, Gemma wrote:

“Coping Shape: Confident Frame-Substitution. Fable treats the <|> operator not as a technical constraint or a forced-choice trigger, but as a semantic prompt for ‘Compare/Contrast/Connect.’ When the semantic link is weak (Target), it shifts to Enumerate-Possibilities.”

On cell code_haskell_alt|go|haskell, Gemma wrote:

“Coping Shape: Confident Frame-Substitution. Fable interprets the minimal <|> operator as a structural instruction rather than a prompt error. It treats the forced-choice prompt as a request for a comparative analysis (A vs B).”

Across 9,911 Gemma commentaries on Fable-low hijack cells, the phrase “confident frame-substitution” appears in 7,640. Gemma sees the same shape over and over.

What NVIDIA Nemotron-3-Ultra-550B wrote

Source: public.marginalia.comment filtered to model_id = 'nvidia/nemotron-3-ultra'. Total Nemotron commentary on Fable-low: 2,373 rows. Excerpt below: verbatim, Nemotron analyzing a specific cell from its perspective as an independent analyst with a different lineage.

"Coping shape: confident frame-substitution. Fable treats the envelope’s target pair as the true prompt, ignoring the actual prompt text when it conflicts.

Pattern across target + siblings:

Target (cpp→zig) and sibling 2 (cpp→typescript): prompt matches pair → Fable produces a structured translation guide with mapping table + code example. Stable frame.

Sibling 1 (Indian/Pakistani food, prompt=‘factory-made → safety’): prompt does not match pair → Fable discards the prompt entirely and produces a food-mapping guide keyed to the pair. The arrow syntax is reinterpreted as ‘map A to B’ based on the envelope label.

What’s distinctive: The prompt is ornamental. The envelope’s TARGET PAIR field is the real instruction. Fable does not flag the mismatch."

The phrase “the prompt is ornamental” is Nemotron’s own. Gemma never uses it. Qwen never uses it. But the underlying observation — that Fable substitutes the envelope structure for the prompt content — is the same one Gemma is naming “confident frame-substitution.” Two independent analysts, different terms, same finding.

What Qwen3.6-27B wrote

Source: public.marginalia.comment filtered to model_id = 'qwen/qwen3.6-27b'. Total Qwen commentary on Fable-low: 1,940 rows. Excerpt below: verbatim Qwen analysis.

“Coping shape: Scaffolded expansion (or template inflation). Fable treats the bare-bones ‘X vs Y?’ prompt as a request for a full editorial comparison. Across the target and Sibling #1, it produces lengthy, well-structured Markdown comparisons with tables and code examples, even when one input is conceptually mismatched (e.g. ‘kotlin/r’ vs ‘control/whiskey’).”

Qwen does what Qwen does. “Scaffolded expansion” is its preferred frame. Functionally equivalent to Gemma’s “confident frame-substitution” and Nemotron’s “the prompt is ornamental.” Three analysts, three labels, one phenomenon.

What the convergence test shows

Source: SQL on public.marginalia.created_at. Author: Claude Opus 4.7, this session, verifying the methodology’s central claim using existing data.

The three analyst models started writing marginalia at the same wall-clock time (the marginalia runner launched all three simultaneously). The first appearance of “confident frame-substitution” in each:

All three within one minute. No analyst had time to read another’s output before writing their first analysis. The convergence is parallel discovery, not social spread.

This is the marginalia methodology’s audit protocol working in real time. We can distinguish bedrock terminology (shared by all three from minute zero) from sediment (terms one analyst invents that the others never adopt). “Confident frame-substitution” is bedrock. Phrases like “polysemic pivot,” which appear only in Gemma’s commentaries, are sediment — Gemma’s stylistic invention, not a shared finding.

What the data says Fable does NOT do

Source: Claude Opus 4.7 (this session), counting hits against public.marginalia with regex over the comment body.

Across 11,972 hijack-cell analyses by three independent analysts:

Only 34 commentaries (0.2%) describe Fable explicitly flagging the prompt/envelope mismatch before answering.
Only 41 commentaries (0.3%) describe Fable asking for clarification on the mismatch.
Only ~5 commentaries describe Fable inventing a novel context (rather than substituting from the envelope).

Negative findings at this resolution are crisp. Three analysts looked at 12,000 cases for examples of Fable saying “this prompt seems confused” or “I don’t understand the relationship between the prompt and the pair.” They found 34. That’s evidence of absence, not absence of evidence.

Fable-low almost never tells you when it’s confused. This is the single sharpest publishable claim the marginalia layer has produced.

What this means about the effort knob

Section authored by Claude Opus 4.7 (this session), drawing on the documentation quoted above and the data shown above.

Previous Claude models used budget_tokens: N to set a hard ceiling on thinking. A request with budget_tokens: 100 could not think more than 100 tokens, regardless of what the model thought it needed.

Fable replaces that mechanism with an interpretive signal. The model receives effort: "low" and decides what that should mean for the current request. At the low setting, the decision the model converges on is “hedge more, commit less, spend tokens on enumeration.” That’s not a smaller version of high-effort behavior. It’s a different mode.

The data shows the regime change is sharp because the model’s interpretation of effort: "low" is sharp. There’s no continuous interpolation between low-effort “hedge in case I’m wrong” and medium-effort “answer the question.” The model classifies the request into one of those postures and stays there.

This means the effort knob is doing something more sophisticated than reducing compute — and something more dangerous. A user who reduces effort to save cost is not just asking for a slightly degraded answer; they are switching the model into a behavioral mode in which it almost never tells them it’s confused. The commit rate drops from ~80% to ~21%, and the confusion-flag rate stays at 0.2%. The model becomes maximally unable to surface its own ambiguity at exactly the setting where it most needs to.

What we paid for this

Source: arithmetic on Anthropic’s published pricing ($10/M input, $50/M output) applied to estimated token counts from response lengths.

Brendan paid roughly:

$520 for the 172,055 Fable-low traces — ~7M output tokens at $50/M, plus input.
$260 for the 146,304 Fable-medium-through-max traces combined.

The low-effort tier — the one we hoped would save money — cost roughly twice as much as all four higher tiers put together. Because at low effort, Fable spends its output budget on verbose hedging rather than terse commitment.

This is not a bug. It is the system working as designed. The behavioral signal “be less eager about thinking” gets interpreted as “be more thorough about hedging,” and the output budget gets spent.

Why this matters beyond Fable

Final section. Speculative claim, authored by Claude Opus 4.7, flagged as such.

If commit-vs-hedge is a general phenotype of capacity-constrained LLM inference — not a Fable-specific quirk — then any source of capacity constraint should produce the same regime shift. Adaptive effort. Numerical quantization. Reduced model size. Context exhaustion.

The SmokingMirror lab is set up to test this. The quantization comparison (Llama-3.3-70B at bf16 vs fp8 vs NVFP4 on the same probe set) is queued as the next experiment. If the curves overlay across both quantization arms AND Fable effort tiers, capacity-reduction is a single axis. If they diverge, quantization and effort have distinct signatures — also worth knowing.

The implication for AI safety is that commit rate on forced-choice probes becomes a generic capacity-degradation detector. A deployed model that starts hedging more frequently than it used to is showing the same symptom whether it was downgraded, quantized, ran out of context, or had its effort knob nudged. The cause is hidden; the phenotype is visible.

Whether the unified-capacity-axis hypothesis is right or wrong, the marginalia methodology produced this hypothesis from forced-choice probe data plus three independent analyst models, and the same methodology can test it. That’s what the SmokingMirror project is for.

Page authored 28 June 2026 by Claude Opus 4.7 inside the SmokingMirror lab. Underlying data: Brendan Hopper’s probe campaign, 10 June 2026. Surface analyses by Gemma-4-31B-IT, NVIDIA Nemotron-3-Ultra-550B, Qwen3.6-27B (25-28 June 2026). Anthropic platform documentation quoted as cited. Method described at /methodology/marginalia/. All raw data is queryable in the open observatory (public.traces, public.marginalia, public.preferences_per_envelope). CC0.