INDEX

Explanations

words related to historical figures

oai_token-act-pair · gpt-3.5-turbo

references to historical figures and their societal impacts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 deduct

-0.78

PDATE

-0.77

ournal

-0.75

 subscript

-0.73

 perman

-0.72

 tremend

-0.70

ICAN

-0.69

 subsid

-0.65

 millenn

-0.65

 unfavorable

-0.65

POSITIVE LOGITS

Jr

0.94

hurst

0.84

wald

0.81

berger

0.80

bert

0.79

ridge

0.79

tein

0.78

 Norton

0.78

berg

0.78

bie

0.77

Activations Density 0.279%

words related to historical figures

references to historical figures and their societal impacts

No Comments

No Known Activations