INDEX

Explanations

sources or references mentioned in a text

oai_token-act-pair · gpt-3.5-turbo

references to sources of information and citations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

Ħ¢

-0.84

aeper

-0.83

eers

-0.69

joined

-0.69

roe

-0.65

eer

-0.65

gran

-0.64

 Potato

-0.61

ey

-0.59

 Jury

-0.57

POSITIVE LOGITS

 sources

1.23

 Sources

1.15

ources

1.14

ourcing

0.94

ource

0.87

Sources

0.86

etter

0.86

source

0.85

 sourcing

0.83

 source

0.82

Activations Density 0.014%

sources or references mentioned in a text

references to sources of information and citations

No Comments

No Known Activations