INDEX

Explanations

terms related to government institutions and military collaboration

oai_token-act-pair · gpt-3.5-turbo

references to government and military entities

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

phies

-0.85

ãĥĻ

-0.70

 VIDEOS

-0.65

rosso

-0.65

tions

-0.65

hots

-0.65

standing

-0.65

models

-0.64

perse

-0.63

bring

-0.63

POSITIVE LOGITS

 itself

0.97

 treasury

0.94

 bureaucracy

0.87

 purse

0.87

 judiciary

0.81

 coffers

0.80

 censor

0.80

 Office

0.77

 embassy

0.76

 Department

0.75

Activations Density 0.320%

terms related to government institutions and military collaboration

references to government and military entities

No Comments

No Known Activations