INDEX

Explanations

occurrences of the word "Note."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.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

luaj

-0.81

ravel

-0.71

 Cosponsors

-0.69

atom

-0.68

 externalToEVAOnly

-0.65

icum

-0.65

aughtered

-0.64

 nomine

-0.64

ULTS

-0.64

 neighb

-0.63

POSITIVE LOGITS

books

1.20

book

1.11

ably

1.08

BOOK

0.98

 caveats

0.74

 note

0.73

ually

0.72

how

0.70

Pad

0.68

pad

0.68

Activations Density 0.024%

occurrences of the word "Note."

No Comments

No Known Activations