INDEX

Explanations

names and places within a text

oai_token-act-pair · gpt-3.5-turbo

names of people and their associated information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

Ire

-0.59

 attRot

-0.59

 toget

-0.57

 adolesc

-0.54

":[

-0.53

 notor

-0.53

 regul

-0.53

 mosqu

-0.52

 recip

-0.50

 nodd

-0.49

POSITIVE LOGITS

escription

0.83

 Cosponsors

0.68

ename

0.57

[-

0.56

 underestimated

0.55

uilding

0.54

 Converted

0.54

omics

0.53

|--

0.52

};

0.52

Activations Density 0.945%

names and places within a text

names of people and their associated information

No Comments

No Known Activations