INDEX

Explanations

terms related to social injustice and oppression

oai_token-act-pair · gpt-3.5-turbo

terms related to oppression and social injustice

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

tein

-0.80

cephal

-0.70

WAYS

-0.70

ocular

-0.67

gran

-0.66

sal

-0.66

edi

-0.65

ãĥ¥

-0.64

crypt

-0.64

cker

-0.63

POSITIVE LOGITS

 oppression

1.04

 oppress

0.96

 oppressed

0.93

 Struggle

0.88

 disadvant

0.86

retched

0.79

ĸļ

0.79

 Palest

0.78

 oppressive

0.75

 prejudice

0.72

Activations Density 0.027%

terms related to social injustice and oppression

terms related to oppression and social injustice

No Comments

No Known Activations