INDEX

Explanations

technical terms related to data and file manipulation

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

CLASS

-0.74

mn

-0.70

rick

-0.69

Shot

-0.68

Thompson

-0.64

ERAL

-0.61

Hung

-0.59

ctor

-0.59

shed

-0.58

OUS

-0.56

POSITIVE LOGITS

 themselves

1.21

 individually

1.03

selves

0.98

 separately

0.96

 consecut

0.95

heet

0.94

 aloud

0.94

etting

0.90

creen

0.89

 manually

0.89

Activations Density 0.502%

technical terms related to data and file manipulation

No Comments

No Known Activations