INDEX

Explanations

references to the character Wonder Woman

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

 destro

-0.82

 manslaughter

-0.77

aution

-0.75

 separat

-0.71

EStream

-0.70

 kindred

-0.70

escription

-0.68

 externalToEVAOnly

-0.66

 seiz

-0.66

 disg

-0.63

POSITIVE LOGITS

bolt

1.09

 Woman

0.93

fully

0.92

power

0.89

stru

0.88

kids

0.88

wall

0.88

storms

0.86

frey

0.86

Woman

0.85

Activations Density 0.021%

references to the character Wonder Woman

No Comments

No Known Activations