INDEX

Explanations

proper nouns and numbers

oai_token-act-pair · gpt-3.5-turbo

specific numerical identifiers and proper names associated with video games and television shows

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ahime

-0.75

cific

-0.74

ryu

-0.73

bleacher

-0.72

tumblr

-0.72

 Chronicle

-0.71

EStreamFrame

-0.70

 guiActiveUnfocused

-0.69

uliffe

-0.69

DragonMagazine

-0.67

POSITIVE LOGITS

·

0.70

 unden

0.70

°

0.68

ikan

0.68

 unpop

0.67

½

0.67

ĺ

0.66

ĩ

0.65

oton

0.64

ī

0.64

Activations Density 0.688%

proper nouns and numbers

specific numerical identifiers and proper names associated with video games and television shows

No Comments

No Known Activations