INDEX

Explanations

phrases that prompt the reader to subscribe or sign up

oai_token-act-pair · gpt-3.5-turbo

repeated uses of the word "our"

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

 Izan

-0.81

bender

-0.78

netflix

-0.74

yang

-0.73

Ã©e

-0.73

lessness

-0.71

stood

-0.71

ivas

-0.71

nesota

-0.71

matter

-0.71

POSITIVE LOGITS

selves

1.01

own

0.93

 newest

0.88

 latest

0.86

 respective

0.84

 sister

0.81

 exclusive

0.81

 motto

0.80

 editorial

0.79

 extensive

0.78

Activations Density 0.068%

phrases that prompt the reader to subscribe or sign up

repeated uses of the word "our"

No Comments

No Known Activations