INDEX

Explanations

mentions of political organizations, government entities, and military actions

oai_token-act-pair · gpt-3.5-turbo

references to government bodies and political events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 prest

-0.64

ãĥ¼ãĥĨ

-0.60

halla

-0.59

ãĥ¼ãĥĨãĤ£

-0.55

$.

-0.53

 Siber

-0.53

eday

-0.52

ecd

-0.51

Occ

-0.51

Liter

-0.51

POSITIVE LOGITS

 reacts

0.83

 approves

0.79

 condemns

0.79

 Says

0.79

 denies

0.78

 warns

0.76

 accuses

0.76

 threatens

0.76

 shuts

0.75

 announces

0.74

Activations Density 0.639%

mentions of political organizations, government entities, and military actions

references to government bodies and political events

No Comments

No Known Activations