INDEX

Explanations

years mentioned in the text

oai_token-act-pair · gpt-3.5-turbo

specific years mentioned in the text

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

EStreamFrame

-0.76

Condition

-0.72

abilia

-0.62

cheat

-0.61

bryce

-0.60

iazep

-0.59

Widget

-0.59

goo

-0.58

ificate

-0.58

 whipping

-0.56

POSITIVE LOGITS

lance

0.72

 Newsweek

0.66

ruary

0.65

ileaks

0.64

 they

0.63

 there

0.62

 Suzuki

0.62

ODUCT

0.62

0.61

 meanwhile

0.61

Activations Density 0.128%

years mentioned in the text

specific years mentioned in the text

No Comments

No Known Activations