INDEX

Explanations

names of individuals and organizations

oai_token-act-pair · gpt-3.5-turbo

titles of movies or shows and references to notable personalities

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

Kik

-0.63

 multiplication

-0.57

ãĥ¼ãĥĨ

-0.55

 scrut

-0.55

 ingred

-0.53

 redundancy

-0.52

 blockers

-0.52

Kop

-0.52

qv

-0.51

ãĤĮ

-0.51

POSITIVE LOGITS

 pione

0.58

 unveiled

0.57

tesy

0.54

TAIN

0.54

zbollah

0.53

iren

0.52

 hailed

0.52

 attracted

0.50

 respectively

0.50

 celebrates

0.50

Activations Density 1.175%

names of individuals and organizations

titles of movies or shows and references to notable personalities

No Known Activations