INDEX

Explanations

decontextualized text with formatting errors

oai_token-act-pair · gpt-3.5-turbo

references to limitations or conditions affecting experiences or situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 Britann

-0.69

Ops

-0.68

 Reloaded

-0.68

 Directory

-0.64

 Examination

-0.64

 penned

-0.61

 Nutr

-0.60

 Scotia

-0.60

 HSBC

-0.60

 gauge

-0.59

POSITIVE LOGITS

too

1.05

great

1.02

selves

0.98

mom

0.98

warm

0.95

same

0.95

dead

0.93

recomm

0.92

wrong

0.90

being

0.90

Activations Density 0.076%

decontextualized text with formatting errors

references to limitations or conditions affecting experiences or situations

No Comments

No Known Activations