INDEX

Explanations

references containing the word "win"

oai_token-act-pair · gpt-3.5-turbo

instances of the word "win," indicating a focus on winning or victories

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

hovah

-0.73

trak

-0.72

erity

-0.68

 reservations

-0.63

rush

-0.61

 Authorization

-0.59

 Retirement

-0.59

 Standing

-0.58

Ĥª

-0.58

 Known

-0.57

POSITIVE LOGITS

ners

1.45

nings

0.97

ning

0.95

ctive

0.89

ner

0.88

NER

0.88

ious

0.87

boats

0.85

win

0.84

iors

0.84

Activations Density 0.014%

references containing the word "win"

instances of the word "win," indicating a focus on winning or victories

No Comments

No Known Activations