INDEX

Explanations

phrases related to professional roles or job descriptions

oai_token-act-pair · gpt-3.5-turbo

phrases related to speeches and formal events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

eters

-0.59

Reviewer

-0.55

esters

-0.55

nces

-0.54

 Firstly

-0.52

 Bought

-0.52

 Originally

-0.52

-)

-0.52

 reacts

-0.52

 lately

-0.51

POSITIVE LOGITS

 eventual

0.69

 ensuing

0.64

 nearby

0.63

ļéĨĴ

0.62

coming

0.61

SPONSORED

0.61

 lieu

0.60

 prest

0.59

$.

0.57

 whom

0.56

Activations Density 0.856%

phrases related to professional roles or job descriptions

phrases related to speeches and formal events

No Comments

No Known Activations