INDEX

Explanations

terms related to police actions, such as arrests and detainments

oai_token-act-pair · gpt-3.5-turbo

terms related to arrests and detentions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ãĤ´ãĥ³

-0.78

ãĤ¡

-0.76

itsch

-0.75

£ı

-0.75

fax

-0.72

TPP

-0.66

vernment

-0.66

POST

-0.65

é¾įåĸļå£«

-0.65

fml

-0.64

POSITIVE LOGITS

him

0.94

 suspects

0.92

Lt

0.79

 dozens

0.78

 fugitive

0.78

 Mohamed

0.77

 individuals

0.76

 former

0.75

llah

0.74

 Wend

0.74

Activations Density 0.150%

terms related to police actions, such as arrests and detainments

terms related to arrests and detentions

No Comments

No Known Activations