INDEX

Explanations

groups of people or entities mentioned in various contexts

oai_token-act-pair · gpt-3.5-turbo

mentions of specific groups of people or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

66666666

-0.59

iasis

-0.59

 arteries

-0.59

 Machines

-0.57

lycer

-0.55

agate

-0.55

 decay

-0.55

dar

-0.54

 continuum

-0.54

 Correspond

-0.53

POSITIVE LOGITS

 member

0.95

 consisting

0.89

members

0.87

 members

0.86

think

0.81

member

0.80

of

0.79

 comprised

0.78

 assembled

0.76

 comprising

0.75

Activations Density 0.103%

groups of people or entities mentioned in various contexts

mentions of specific groups of people or entities

No Comments

No Known Activations