INDEX

Explanations

proper nouns and names

oai_token-act-pair · gpt-3.5-turbo

references to identification or personal details

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

naissance

-0.62

 Annotations

-0.61

PDATE

-0.61

atten

-0.59

prising

-0.57

senal

-0.56

-0.55

cffff

-0.54

attering

-0.54

 whichever

-0.53

POSITIVE LOGITS

is

1.33

is

1.15

IS

1.02

Is

0.99

isn

0.95

are

0.93

was

0.91

are

0.80

isin

0.77

Is

0.77

Activations Density 0.241%

proper nouns and names

references to identification or personal details

No Comments

No Known Activations