INDEX

Explanations

instances where something is officially excluded as a possibility or decision is made

oai_token-act-pair · gpt-3.5-turbo

words related to decisions or authoritative statements

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

issance

-0.86

akra

-0.82

velength

-0.75

vas

-0.73

rious

-0.72

ufact

-0.72

ription

-0.71

illin

-0.70

ãĥ¼ãĥĨãĤ£

-0.70

ãĥ¤

-0.64

POSITIVE LOGITS

phas

0.76

 differently

0.75

out

0.73

 decisively

0.73

 against

0.69

 unanimously

0.68

 definitively

0.68

doms

0.66

 apart

0.65

maker

0.64

Activations Density 0.028%

instances where something is officially excluded as a possibility or decision is made

words related to decisions or authoritative statements

No Comments

No Known Activations