INDEX

Explanations

violent actions or descriptions

oai_token-act-pair · gpt-3.5-turbo

instances of violence or physical harm

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ivals

-0.77

enario

-0.76

 Franch

-0.75

 Families

-0.74

 Infrastructure

-0.73

 ourselves

-0.71

 Income

-0.70

levard

-0.68

edia

-0.68

 Skies

-0.66

POSITIVE LOGITS

 numb

1.06

 protr

0.94

stretched

0.92

 amput

0.91

 swollen

0.88

 throb

0.87

 clenched

0.86

 abnorm

0.80

 shaved

0.80

 curled

0.77

Activations Density 0.331%

violent actions or descriptions

instances of violence or physical harm

No Comments

No Known Activations