INDEX

Explanations

words related to television shows and movies

oai_token-act-pair · gpt-3.5-turbo

references to comparisons, particularly using the word "like."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.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

ccording

-0.79

essing

-0.74

ennes

-0.70

ittal

-0.70

uliffe

-0.70

igious

-0.70

atform

-0.70

ackers

-0.69

essed

-0.68

acia

-0.68

POSITIVE LOGITS

lihood

1.43

 ours

0.94

liest

0.92

 hers

0.91

lier

0.83

 yours

0.77

unts

0.75

those

0.73

 those

0.72

ãĥ©ãĥ³

0.68

Activations Density 0.063%

words related to television shows and movies

references to comparisons, particularly using the word "like."

No Comments

No Known Activations