INDEX

Explanations

phrases related to entry or the act of entering

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "the" often in various contexts

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

Not in Any Lists

No Comments

Negative Logits

rand

-0.81

each

-0.76

worn

-0.73

abel

-0.69

aeus

-0.69

 survives

-0.67

arians

-0.66

udder

-0.65

handedly

-0.65

 located

-0.63

POSITIVE LOGITS

 fray

1.41

 confines

1.03

 same

0.98

 realm

0.96

 forefront

0.92

 city

0.89

 ranks

0.88

 marketplace

0.88

 halls

0.87

 midst

0.86

Activations Density 0.383%

phrases related to entry or the act of entering

mentions of the word "the" often in various contexts

No Comments

No Known Activations