INDEX

Explanations

phrases related to specific titles, terms, or brands mentioned in the text

oai_token-act-pair · gpt-3.5-turbo

phrases related to health and medical conditions

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 sped

-0.59

 dashed

-0.58

 conserv

-0.57

anwhile

-0.56

 referen

-0.54

heastern

-0.54

 fortified

-0.54

 accelerated

-0.52

 protected

-0.52

 complementary

-0.52

POSITIVE LOGITS

 Syndrome

0.89

opia

0.80

Fest

0.78

 Productions

0.72

Wiki

0.72

topia

0.67

ventures

0.66

Mania

0.66

Fit

0.65

Gate

0.64

Activations Density 0.778%

phrases related to specific titles, terms, or brands mentioned in the text

phrases related to health and medical conditions

No Known Activations