INDEX

Explanations

phrases describing someone or something being away or absent

oai_token-act-pair · gpt-3.5-turbo

instances of the word "out."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

asus

-0.70

cius

-0.70

uyomi

-0.69

ugh

-0.68

hess

-0.68

hedral

-0.65

Export

-0.64

antry

-0.63

iosity

-0.63

esa

-0.63

POSITIVE LOGITS

fitted

1.21

stretched

1.06

raged

0.94

smart

0.92

doing

0.90

lier

0.88

fitting

0.85

casts

0.85

skirts

0.83

bound

0.82

Activations Density 0.039%

phrases describing someone or something being away or absent

instances of the word "out."

No Comments

No Known Activations