INDEX

Explanations

prepositions indicating origin or direction

oai_token-act-pair · gpt-3.5-turbo

quantitative data and statistics within the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Farage

-0.64

 Canaver

-0.59

 Wenger

-0.57

 Loft

-0.57

 Roose

-0.56

 Slater

-0.56

 Rahul

-0.55

 Whedon

-0.55

Gou

-0.55

 Klopp

-0.55

POSITIVE LOGITS

ãĤ©

0.86

APTER

0.77

Û

0.77

ãĢ

0.73

Beginning

0.72

ãĥŃ

0.71

Â«

0.70

eteen

0.70

irteen

0.69

ornings

0.69

Activations Density 0.874%

prepositions indicating origin or direction

quantitative data and statistics within the text

No Comments

No Known Activations