INDEX

Explanations

phrases related to matching or aligning things to each other

oai_token-act-pair · gpt-3.5-turbo

instances of the word "correspond" and its variations

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

uv

-0.82

Downloadha

-0.71

oult

-0.70

 helicop

-0.70

cloth

-0.68

 Bedford

-0.66

 Scotia

-0.65

selection

-0.65

ils

-0.63

arde

-0.63

POSITIVE LOGITS

ingly

0.89

acters

0.82

ioned

0.76

ificantly

0.75

 newsp

0.74

ivalent

0.70

 corresponds

0.69

MPG

0.69

alties

0.68

 correspond

0.68

Activations Density 0.011%

phrases related to matching or aligning things to each other

instances of the word "correspond" and its variations

No Comments

No Known Activations