INDEX

Explanations

references to articles or written content

oai_token-act-pair · gpt-3.5-turbo

references to articles or pages

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

seys

-0.68

sbm

-0.68

sed

-0.66

 warranties

-0.65

 selves

-0.62

 Maid

-0.62

 stripes

-0.61

 oneself

-0.61

abl

-0.61

 speeches

-0.61

POSITIVE LOGITS

 adapted

0.74

 grate

0.66

 appl

0.64

adapt

0.62

ittee

0.62

rep

0.61

gha

0.61

REC

0.60

ground

0.59

land

0.59

Activations Density 0.074%

references to articles or written content

references to articles or pages

No Comments

No Known Activations