Claim · Date unknown · Stated in source · Supported

Claude 3.5 Sonnet Model Card Addendum states what the model-card question “Stated limits” asks.

One recorded claim. The claim panel gives its qualification, the source passages with their place in each source, and how it was checked.

What it connects

Claude 3.5 Sonnet Model Card Addendum → states provision → Stated limits

Where it appears

Model cards compared →

Claim and evidence

Claude 3.5 Sonnet Model Card Addendumstates provision ↓Stated limits

Machine-checkedStated in sourceSupported

Supporting evidence: primary · 1 source

Provision: Model-card question · provision ID model-card-q16
Date: Date unknown

Scope and limitsThe addendum states that safety training can lead capability evaluations to underestimate underlying capability, and that its results may do so. It also states that much remains to be learned about evaluation science and cadence. What the card states, quoted from the expression read. A card is the developer’s own account: not evidence that an evaluation was run as described, that a safeguard works, or that a finding is independent.

The evidence 01

primary sourceSupports the claim

Claude 3.5 Sonnet Model Card Addendum

Anthropic · Publication date unknown

Archived copy, captured 2024-06-20: Claude 3.5 Sonnet Model Card Addendum

Cite the original source

Cite the original document, not Selekoa. Check these details against it.

Source excerpt

“Note that it is possible that the results underestimate the capability of an equivalent Helpful-only model, because alignment training may cause it to hedge concerning answers.”

Location in source Section 3, 'Safety Evaluations Overview', PDF p. 6

How it was read Retrieved directly from the publisher

Machine-checked · 2026-11-04

Source details and check notes

Authors: Anthropic
Version: Retrieved for the candidate pack on 2026-11-04 (SHA-256 ac396c7f155dcbfb33ceb342b48647e330ca518c01c7f508001d6aad93257a03) directly from the publisher’s host. Retrieved again on 2026-11-03 with the same SHA-256. The addendum prints no date. Anthropic's card index gives June 2024, and an update line of the Claude 3 card dates the addendum's addition to that card to June 20, 2024; neither is the addendum's own date. The archive capture was found by an index search of the Internet Archive on 2026-11-03; it was not opened, and its bytes are not authenticated.
Canonical address: Open the canonical source ↗
Accessed: 2026-11-04
Language: English (en)
Archive copy: Open archived copy, captured 2024-06-20: Claude 3.5 Sonnet Model Card Addendum. No publication date is recorded for this source. An archive copy shows only what the archive retrieved on 2024-06-20. It does not prove what the page said earlier, and it can differ from the copy checked on 2026-11-04.
Attribution: Anthropic, “Claude 3.5 Sonnet Model Card Addendum”.
Reuse: No open redistribution licence asserted. Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution.
Check notes: Candidate pack passage, model cards compared (research/model-cards-2). The quotation (26 words) matched the card’s text retrieved for the candidate pack on 2026-11-04 (SHA-256 ac396c7f155dcbfb33ceb342b48647e330ca518c01c7f508001d6aad93257a03) and again on 2026-11-04, after the adjudication, in a fresh retrieval with the same SHA-256. Matching ignored differences of whitespace, case, quotation marks, dashes, PDF line-break hyphens and Unicode compatibility forms, so the source’s typography can differ. No human editorial review.

Status and meaning

Editorial status. Machine-checked. Machines found this record and checked it against its sources; no person accepted it individually. Each evidence record above shows how it was checked.

Relation. An exact framework edition, or another document with an exact expression, states one provision: a capability threshold or risk level and its trigger, a required safeguard or action, a decision authority (who decides, recommends, can overrule or is informed), a stated exception, or another statement of its content, such as a licence term or a prospectus statement. The passage records what the document says, not implementation, compliance or effectiveness. Provisions with the same key align across editions of one family.

Date. The card prints no date of its own, so its cells carry no date.

Record ID mc-cell-anthropic-claude-3-5-sonnet-q16