INDEX

Explanations

words related to gaining success or progress

oai_token-act-pair · gpt-3.5-turbo

terms related to traction and gaining influence or support

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

tein

-0.74

 Peninsula

-0.71

 halves

-0.66

ulk

-0.63

rug

-0.62

Que

-0.62

thus

-0.62

que

-0.62

humane

-0.62

cott

-0.59

POSITIVE LOGITS

 traction

1.05

ability

0.83

é¾įå¥ĳå£«

0.76

irtual

0.74

ãĤ¤ãĥĪ

0.73

ibility

0.72

ãĥ¼ãĥĨ

0.72

atility

0.72

oday

0.71

culosis

0.68

Activations Density 0.011%

words related to gaining success or progress

terms related to traction and gaining influence or support

No Comments

No Known Activations