INDEX

Explanations

questions and interrogations

oai_token-act-pair · gpt-3.5-turbo

questions and inquiries

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

apers

-0.76

raped

-0.74

apo

-0.65

 gall

-0.63

memory

-0.62

gered

-0.61

tailed

-0.61

itol

-0.61

clad

-0.61

sels

-0.61

POSITIVE LOGITS

 GOODMAN

0.92

Interview

0.83

laughs

0.76

DK

0.76

Laughs

0.74

 Interview

0.73

 interviews

0.72

Rh

0.69

Fro

0.66

 interview

0.66

Activations Density 0.293%

questions and interrogations

questions and inquiries

No Comments

No Known Activations