INDEX

Explanations

verbs related to actions or behaviors

oai_token-act-pair · gpt-3.5-turbo

references to serious crimes or allegations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 unison

-0.78

selves

-0.65

 respectively

-0.63

angular

-0.59

VERTISEMENT

-0.57

were

-0.56

Ware

-0.56

 constituted

-0.55

 respective

-0.55

merce

-0.54

POSITIVE LOGITS

 himself

1.51

 Himself

1.15

his

1.11

 herself

1.07

HIS

0.91

his

0.85

His

0.83

His

0.78

 reelection

0.68

 tweeting

0.66

Activations Density 1.514%

verbs related to actions or behaviors

references to serious crimes or allegations

No Comments

No Known Activations