INDEX

Explanations

numerous instances of text mentioning submitting letters to editors

oai_token-act-pair · gpt-3.5-turbo

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

Ĥª

-0.61

 indo

-0.58

MV

-0.57

MN

-0.56

IENT

-0.56

 Boys

-0.54

PASS

-0.54

 Mund

-0.54

------------------------------------------------

-0.53

 Cind

-0.53

POSITIVE LOGITS

letters

0.72

Journal

0.68

ibrarian

0.64

 inacc

0.62

Correct

0.62

Correction

0.62

flat

0.62

gha

0.61

mbudsman

0.60

 editors

0.60

Activations Density 0.071%

numerous instances of text mentioning submitting letters to editors

No Comments

No Known Activations