INDEX

Explanations

statements or quotes made by individuals

oai_token-act-pair · gpt-3.5-turbo

statements that assess the reliability or credibility of sources in discourse

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ascus

-0.60

rats

-0.59

Contents

-0.59

thur

-0.59

steamapps

-0.58

iosity

-0.56

successfully

-0.55

 diaper

-0.54

opus

-0.53

eeds

-0.53

POSITIVE LOGITS

 said

0.70

said

0.70

 explained

0.69

 countered

0.68

 spokesman

0.66

 analyst

0.66

 spokeswoman

0.64

 lamented

0.63

 Fried

0.62

 analysts

0.62

Activations Density 0.372%

statements or quotes made by individuals

statements that assess the reliability or credibility of sources in discourse

No Comments

No Known Activations