INDEX

Explanations

phrases indicating the significance or importance of certain actions or items

oai_token-act-pair · gpt-3.5-turbo

terms that emphasize the significance or necessity of various concepts or actions

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

 itch

-0.61

eely

-0.61

riage

-0.58

Hell

-0.57

Snow

-0.56

åħī

-0.56

 fever

-0.56

bugs

-0.56

 WARRANT

-0.55

ryan

-0.54

POSITIVE LOGITS

 nonetheless

0.83

nell

0.83

 enough

0.82

entimes

0.74

ially

0.72

arily

0.71

 adjunct

0.68

 distingu

0.67

 ingred

0.67

 proble

0.67

Activations Density 0.104%

phrases indicating the significance or importance of certain actions or items

terms that emphasize the significance or necessity of various concepts or actions

No Comments

No Known Activations