INDEX

Explanations

instances of the word "walking"

oai_token-act-pair · gpt-3.5-turbo

instances of the word "walking."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

Nev

-0.74

ucks

-0.74

iled

-0.72

alin

-0.71

nces

-0.71

ancies

-0.70

oil

-0.68

ARA

-0.67

aucuses

-0.67

urities

-0.64

POSITIVE LOGITS

walking

0.95

 walking

0.89

walk

0.81

walker

0.81

 Walk

0.80

Walk

0.78

 stride

0.78

 Walking

0.77

 walk

0.76

 upright

0.76

Activations Density 0.007%

instances of the word "walking"

instances of the word "walking."

No Comments

No Known Activations