INDEX

Explanations

proper nouns, specifically names of people or places

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ivity

-0.69

 Rivals

-0.69

ivities

-0.66

IFIED

-0.65

assets

-0.64

atever

-0.62

atform

-0.61

iffs

-0.60

ITY

-0.60

ively

-0.60

POSITIVE LOGITS

leck

0.81

enthal

0.78

erm

0.76

elf

0.76

hal

0.75

eer

0.75

ering

0.74

0.73

erman

0.72

pend

0.72

Activations Density 0.044%

proper nouns, specifically names of people or places

No Comments

No Known Activations