INDEX

Explanations

words related to specific professions or fields, e.g., doctoring, weatherporn

oai_token-act-pair · gpt-3.5-turbo

terms related to specific forms of communication or entertainment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

shall

-0.64

OPA

-0.60

BILITIES

-0.59

 Wein

-0.58

inelli

-0.56

Ĥİ

-0.55

NAS

-0.53

SAN

-0.52

RH

-0.52

INTON

-0.52

POSITIVE LOGITS

tainment

0.79

aneous

0.70

ously

0.63

idepress

0.63

usterity

0.62

aneously

0.62

itory

0.60

acters

0.59

aries

0.59

naires

0.58

Activations Density 0.779%

words related to specific professions or fields, e.g., doctoring, weatherporn

terms related to specific forms of communication or entertainment

No Comments

No Known Activations