INDEX

Explanations

the names of authors in various fields

oai_token-act-pair · gpt-3.5-turbo

references to authors and their works

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ooth

-0.70

asia

-0.70

feat

-0.68

 specificity

-0.64

urden

-0.61

 uncond

-0.60

tone

-0.57

 palate

-0.56

 insign

-0.56

 propri

-0.56

POSITIVE LOGITS

 Bearing

0.69

@#&

0.69

thood

0.68

Initialized

0.68

 RELE

0.68

olkien

0.65

olls

0.64

ELY

0.64

stellar

0.63

[|

0.62

Activations Density 0.112%

the names of authors in various fields

references to authors and their works

No Comments

No Known Activations