INDEX

Explanations

words related to political power or authority

oai_token-act-pair · gpt-3.5-turbo

discussions about power dynamics and authority

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

Von

-0.71

eday

-0.70

eret

-0.68

riad

-0.68

TAG

-0.66

ript

-0.66

 Taste

-0.64

IGH

-0.63

ALK

-0.62

algia

-0.62

POSITIVE LOGITS

lessness

1.06

houses

1.04

stroke

0.97

 vested

0.95

FUL

0.90

 wielded

0.89

 levers

0.88

holder

0.88

lifting

0.84

puff

0.82

Activations Density 0.039%

words related to political power or authority

discussions about power dynamics and authority

No Comments

No Known Activations