INDEX

Explanations

names of individuals and companies

oai_token-act-pair · gpt-3.5-turbo

names of notable figures or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 withd

-0.58

lik

-0.58

 cred

-0.55

 answ

-0.55

":"/

-0.55

 rather

-0.54

 appre

-0.54

 itself

-0.54

Ö¼

-0.53

 situ

-0.53

POSITIVE LOGITS

etc

0.88

ĪĴ

0.86

0.74

âĢ¦)

0.74

etc

0.70

,...

0.70

 Together

0.67

Together

0.66

...)

0.65

 Sisters

0.63

Activations Density 0.236%

names of individuals and companies

names of notable figures or entities

No Comments

No Known Activations