INDEX

Explanations

mentions of processed meats or foods

oai_token-act-pair · gpt-3.5-turbo

instances of the word "processed" and related contexts

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

ansky

-0.68

fall

-0.67

bid

-0.66

vae

-0.65

 Lazarus

-0.64

bors

-0.63

 contrasting

-0.62

 passionate

-0.62

lust

-0.60

 shining

-0.59

POSITIVE LOGITS

 processed

1.05

 Processing

0.91

 processing

0.87

ichick

0.85

atoes

0.80

imore

0.80

chnology

0.80

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

0.79

Initialized

0.79

avascript

0.78

Activations Density 0.013%

mentions of processed meats or foods

instances of the word "processed" and related contexts

No Comments

No Known Activations