INDEX

Explanations

short phrases or sentences starting with "No" and followed by a verb or a description

oai_token-act-pair · gpt-3.5-turbo

instances of the word "No."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ãĤ¼ãĤ¦ãĤ¹

-0.78

RAFT

-0.78

rosse

-0.78

assies

-0.75

otle

-0.70

iership

-0.70

tein

-0.70

lycer

-0.69

aven

-0.67

 tong

-0.66

POSITIVE LOGITS

etheless

1.03

zzle

0.89

xious

0.88

except

0.84

AH

0.84

 longer

0.83

ct

0.83

terday

0.82

obs

0.81

emi

0.78

Activations Density 0.023%

short phrases or sentences starting with "No" and followed by a verb or a description

instances of the word "No."

No Comments

No Known Activations