INDEX

Explanations

information or announcements that are being presented or shared

oai_token-act-pair · gpt-3.5-turbo

instances of the word "here."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 Strait

-0.65

acci

-0.62

)].

-0.59

 Breath

-0.58

 AIDS

-0.57

Edge

-0.57

arij

-0.56

 Heights

-0.55

Seg

-0.55

Sin

-0.55

POSITIVE LOGITS

tical

1.27

tics

1.24

abouts

1.17

tic

1.05

upon

0.81

oys

0.73

inel

0.73

with

0.72

landish

0.72

fires

0.71

Activations Density 0.039%

information or announcements that are being presented or shared

instances of the word "here."

No Comments

No Known Activations