INDEX

Explanations

references to political and economic topics, particularly related to actions and decisions

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

negative or prohibitive expressions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 12-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_post

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

wolves

-0.71

RD

-0.60

 Numbers

-0.59

 Calls

-0.57

 Equality

-0.57

URES

-0.57

 Downs

-0.57

laws

-0.56

 Shelley

-0.56

 aloud

-0.55

POSITIVE LOGITS

iver

1.01

 pione

0.92

0.84

 conflic

0.83

itud

0.80

ivers

0.79

 exting

0.78

icator

0.76

icate

0.74

vertisement

0.74

Activations Density 0.027%

references to political and economic topics, particularly related to actions and decisions

negative or prohibitive expressions

No Comments

No Known Activations