INDEX

Explanations

phrases indicating a source of information or attribution

oai_token-act-pair · gpt-3.5-turbo

references to sources or reports in the text

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

estern

-0.76

otin

-0.68

blast

-0.68

apons

-0.67

igers

-0.67

obyl

-0.67

pez

-0.65

llular

-0.64

 realise

-0.64

aden

-0.63

POSITIVE LOGITS

ly

0.80

Ĥİ

0.78

 sources

0.75

 Sources

0.75

Sources

0.75

Rank

0.75

edly

0.74

translation

0.70

views

0.69

eous

0.68

Activations Density 0.045%

phrases indicating a source of information or attribution

references to sources or reports in the text

No Comments

No Known Activations