INDEX

Explanations

words related to legal proceedings and news reports

oai_token-act-pair · gpt-3.5-turbo

specific symbols representing citations or formal referencing in the text

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

 monop

-0.77

 tremend

-0.75

bda

-0.70

 fart

-0.69

 zoning

-0.68

 Tsuk

-0.65

 Banana

-0.64

 scatter

-0.64

 juggling

-0.64

rons

-0.63

POSITIVE LOGITS

£

0.79

Hon

0.74

¬

0.73

said

0.73

Â¯

0.72

 Rapids

0.70

Â¢

0.69

º

0.69

âĹ¼

0.69

Govern

0.68

Activations Density 0.313%

words related to legal proceedings and news reports

specific symbols representing citations or formal referencing in the text

No Comments

No Known Activations