INDEX

Explanations

mentions of different languages or language-related terms

oai_token-act-pair · gpt-3.5-turbo

references to different languages and their usage

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

arious

-0.86

urion

-0.86

ocrates

-0.86

-0.82

rotein

-0.79

arching

-0.78

ptions

-0.76

Tesla

-0.76

arcity

-0.74

uristic

-0.73

POSITIVE LOGITS

 equivalents

0.83

 learners

0.81

 proficiency

0.81

 bilingual

0.79

 immersion

0.76

 supremacy

0.76

 dictionary

0.75

 spoken

0.75

 interpre

0.73

 hegemony

0.73

Activations Density 0.043%

mentions of different languages or language-related terms

references to different languages and their usage

No Comments

No Known Activations