INDEX

Explanations

words related to social issues or societal observations

oai_token-act-pair · gpt-3.5-turbo

terms related to social values and societal issues

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

 Coverage

-0.77

 Bullets

-0.76

heny

-0.74

osate

-0.73

mares

-0.72

edly

-0.69

etics

-0.69

 Volcano

-0.68

 Tours

-0.67

endez

-0.66

POSITIVE LOGITS

oder

0.79

 minded

0.76

folk

0.72

FN

0.72

ering

0.70

onge

0.69

 Skinner

0.68

 Musk

0.68

kun

0.68

agers

0.68

Activations Density 0.026%

words related to social issues or societal observations

terms related to social values and societal issues

No Comments

No Known Activations