INDEX

Explanations

mention of evidence

oai_token-act-pair · gpt-3.5-turbo

references to the concept of evidence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

skill

-0.74

nas

-0.71

nee

-0.66

ija

-0.66

frey

-0.65

care

-0.62

jc

-0.61

 dear

-0.60

awar

-0.60

Care

-0.60

POSITIVE LOGITS

evidence

1.28

 evidence

1.27

 Evidence

1.18

Evidence

1.00

 evid

0.95

 proof

0.89

uments

0.86

proof

0.83

 proofs

0.81

andum

0.81

Activations Density 0.024%

mention of evidence

references to the concept of evidence

No Comments

No Known Activations