INDEX

Explanations

phrases related to decisions, commitments, and positions

oai_token-act-pair · gpt-3.5-turbo

forms of the verb "to stick."

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

=-=-=-=-=-=-=-=-

-0.84

catentry

-0.82

UTF

-0.77

tnc

-0.75

 DRAG

-0.73

cffffcc

-0.69

 GOODMAN

-0.63

OOD

-0.62

tumblr

-0.61

Ã£

-0.61

POSITIVE LOGITS

holders

1.16

holder

1.02

pole

0.79

mate

0.74

helm

0.72

overs

0.72

lishes

0.67

ffen

0.64

warts

0.64

finding

0.64

Activations Density 0.036%

phrases related to decisions, commitments, and positions

forms of the verb "to stick."

No Comments

No Known Activations