INDEX

Explanations

highlights of individuals or personalities

oai_token-act-pair · gpt-3.5-turbo

repeated syllables in words or phrases

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ICLE

-0.59

estial

-0.59

ibilities

-0.58

iop

-0.57

uo

-0.57

abouts

-0.55

idon

-0.55

iosity

-0.54

uten

-0.54

ometric

-0.53

POSITIVE LOGITS

ffer

0.63

 Brach

0.61

FG

0.61

ĵĺ

0.59

ffe

0.59

FN

0.59

kefeller

0.58

legate

0.57

Var

0.57

æ©

0.55

Activations Density 0.127%

highlights of individuals or personalities

repeated syllables in words or phrases

No Comments

No Known Activations