INDEX

Explanations

phrases that talk about comparing or contrasting different experiences

oai_token-act-pair · gpt-3.5-turbo

concepts related to the relationship between race and slavery

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

idth

-0.84

utherland

-0.80

 Cosponsors

-0.78

uve

-0.74

vous

-0.73

arna

-0.72

itely

-0.72

ificantly

-0.71

oulos

-0.70

anmar

-0.69

POSITIVE LOGITS

 damned

0.85

ŃĶ

0.78

DEM

0.70

 damn

0.64

 faith

0.63

rot

0.59

 Alic

0.59

iom

0.59

spirit

0.58

 unex

0.58

Activations Density 0.000%

phrases that talk about comparing or contrasting different experiences

concepts related to the relationship between race and slavery

No Comments

No Known Activations