INDEX

Explanations

references to actions or events in a formal or informative context

oai_token-act-pair · gpt-3.5-turbo

instances of opinion or approval in discussions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

amental

-0.72

ament

-0.68

Ò

-0.68

/+

-0.66

oplan

-0.65

ãĥ´ãĤ¡

-0.64

unique

-0.64

reason

-0.63

oubt

-0.63

rets

-0.62

POSITIVE LOGITS

terday

0.70

AAP

0.68

 devised

0.64

 invited

0.60

 decided

0.59

 curfew

0.59

PB

0.57

fully

0.57

 issuing

0.56

 inscribed

0.56

Activations Density 0.619%

references to actions or events in a formal or informative context

instances of opinion or approval in discussions

No Comments

No Known Activations