INDEX

Explanations

instances of the word "No" followed by a statement

oai_token-act-pair · gpt-3.5-turbo

negations or refusals

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

iership

-0.80

ials

-0.71

lycer

-0.71

RAFT

-0.70

ãĤ¼ãĤ¦ãĤ¹

-0.69

aven

-0.68

romy

-0.67

assies

-0.66

rosse

-0.64

rex

-0.63

POSITIVE LOGITS

xious

1.08

etheless

0.98

 longer

0.95

 matter

0.95

 doubt

0.91

zzle

0.89

terday

0.86

 indication

0.85

 clue

0.84

except

0.83

Activations Density 0.091%

instances of the word "No" followed by a statement

negations or refusals

No Comments

No Known Activations