INDEX

Explanations

the word "turn" with high activation values

oai_token-act-pair · gpt-3.5-turbo

instances of the word "turn" 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 @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

è¦ļéĨĴ

-0.90

capacity

-0.70

foundation

-0.64

ropolitan

-0.63

 Reasons

-0.63

inately

-0.62

itsch

-0.61

 Limits

-0.60

 Opera

-0.60

necess

-0.60

POSITIVE LOGITS

coat

0.81

agra

0.78

overs

0.74

turn

0.73

edit

0.73

about

0.71

wheel

0.70

terday

0.69

atile

0.69

plates

0.67

Activations Density 0.034%

the word "turn" with high activation values

instances of the word "turn" in various contexts

No Comments

No Known Activations