INDEX

Explanations

news agency names

oai_token-act-pair · gpt-3.5-turbo

references to the news organization Reuters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

gran

-0.80

ysis

-0.72

icular

-0.71

alities

-0.69

iencies

-0.69

gone

-0.63

 constants

-0.62

Redditor

-0.61

oven

-0.59

flo

-0.58

POSITIVE LOGITS

Reuters

0.87

Money

0.77

News

0.76

ournal

0.73

+)

0.73

Associated

0.72

)—

0.72

PLIED

0.71

agascar

0.71

CF

0.71

Activations Density 0.006%

news agency names

references to the news organization Reuters

No Comments

No Known Activations