INDEX

Explanations

phrases related to actions or occurrences

oai_token-act-pair · gpt-3.5-turbo

words or phrases indicating conditions or circumstances related to time and causation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

edIn

-0.70

emale

-0.70

ilaterally

-0.68

farious

-0.63

 Seym

-0.62

 helicop

-0.62

akespe

-0.59

 Vaugh

-0.59

 vulner

-0.58

Ire

-0.57

POSITIVE LOGITS

Psy

0.56

 Vulkan

0.54

 Coverage

0.51

 Squirrel

0.51

 Mars

0.49

 ROCK

0.49

 delay

0.49

 Steam

0.49

 DARK

0.48

 Disclosure

0.48

Activations Density 0.486%

phrases related to actions or occurrences

words or phrases indicating conditions or circumstances related to time and causation

No Comments

No Known Activations