INDEX

Explanations

terms related to menu items

oai_token-act-pair · gpt-3.5-turbo

references to menu options

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

å¥³

-0.84

ioxide

-0.72

âĢ¢âĢ¢âĢ¢âĢ¢

-0.68

 Fundamental

-0.67

kj

-0.66

 Sidd

-0.65

 Mary

-0.63

esville

-0.62

theless

-0.62

Coy

-0.61

POSITIVE LOGITS

 menus

0.96

isine

0.96

 menu

0.95

 options

0.87

 Menu

0.84

eteria

0.83

ropri

0.83

uration

0.80

menu

0.80

entry

0.79

Activations Density 0.012%

terms related to menu items

references to menu options

No Comments

No Known Activations