INDEX

Explanations

references to scientific papers

oai_token-act-pair · gpt-3.5-turbo

mentions of academic papers and their authors

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

cffff

-0.71

 Staples

-0.64

 customs

-0.63

cffffcc

-0.62

Ģ

-0.61

ichita

-0.60

 granite

-0.59

 Lansing

-0.59

 Harmony

-0.59

 discretionary

-0.58

POSITIVE LOGITS

papers

0.95

clip

0.87

velop

0.84

uscript

0.83

itatively

0.82

doctoral

0.80

 titled

0.80

Paper

0.79

 published

0.78

 outlining

0.78

Activations Density 0.033%

references to scientific papers

mentions of academic papers and their authors

No Comments

No Known Activations