INDEX

Explanations

instances of non-responses or lack of comments in given contexts

oai_token-act-pair · gpt-3.5-turbo

phrases related to comments or statements made about events or situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

Eva

-0.69

Medium

-0.58

eele

-0.58

ãĥ©ãĥ³

-0.55

Shell

-0.55

fan

-0.54

çĶ°

-0.54

 Penguin

-0.53

 Booster

-0.53

Flickr

-0.53

POSITIVE LOGITS

 anymore

1.72

nor

1.66

nor

1.14

 slightest

1.12

yet

1.00

 whatsoever

0.99

 either

0.89

yet

0.85

 anywhere

0.83

 unless

0.83

Activations Density 1.411%

instances of non-responses or lack of comments in given contexts

phrases related to comments or statements made about events or situations

No Comments

No Known Activations