INDEX

Explanations

phrases indicating comparison or equivalence

oai_token-act-pair · gpt-3.5-turbo

repetitions of the word "just" across various contexts

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

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

Negative Logits

 sidx

-0.70

 seiz

-0.66

>>>>>>>>

-0.59

ership

-0.59

 Also

-0.59

 trave

-0.57

 Cooperation

-0.56

seek

-0.55

apolis

-0.55

 necks

-0.54

POSITIVE LOGITS

ifiable

1.17

ifications

1.08

if

0.92

ices

0.91

ified

0.89

IFIED

0.83

icia

0.83

 desserts

0.79

shy

0.78

 fine

0.78

Activations Density 0.081%

phrases indicating comparison or equivalence

repetitions of the word "just" across various contexts

No Known Activations