INDEX

Explanations

words related to governmental or bureaucratic entities, particularly the State Department

oai_token-act-pair · gpt-3.5-turbo

references to the State Department

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

poons

-0.71

vre

-0.70

 statically

-0.69

eston

-0.68

ãĤ¡

-0.66

enegger

-0.65

razil

-0.65

ateurs

-0.65

urous

-0.64

iously

-0.63

POSITIVE LOGITS

 Dept

1.13

 Department

1.13

 Patrol

0.87

 Secretary

0.86

 Farm

0.84

 Legislature

0.84

wide

0.83

 Legisl

0.83

 University

0.82

 Pages

0.81

Activations Density 0.036%

words related to governmental or bureaucratic entities, particularly the State Department

references to the State Department

No Comments

No Known Activations