INDEX

Explanations

words related to locations and controversial events

oai_token-act-pair · gpt-3.5-turbo

mentions of incidents related to violence or crime

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

pse

-0.90

 UNCLASSIFIED

-0.81

.).

-0.80

 tremend

-0.78

CONCLUS

-0.74

 incorpor

-0.73

 Azerb

-0.73

 unden

-0.72

âĹ¼

-0.70

)."

-0.70

POSITIVE LOGITS

 Belfast

0.75

 Donald

0.68

utterstock

0.66

 ÂŃ

0.64

 allegedly

0.59

 transgender

0.58

 Adolf

0.55

 Hillary

0.55

 frontman

0.55

ÂŃ

0.53

Activations Density 1.499%

words related to locations and controversial events

mentions of incidents related to violence or crime

No Comments

No Known Activations