INDEX

Explanations

proper nouns related to various people or entities

oai_token-act-pair · gpt-3.5-turbo

names or terms related to specific entities or organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

merce

-0.77

 SERV

-0.69

 thous

-0.67

 Mehran

-0.65

 Leban

-0.64

llah

-0.63

 INCLUD

-0.63

perty

-0.63

 Maul

-0.62

 retaliation

-0.62

POSITIVE LOGITS

ibrary

0.91

heed

0.89

gow

0.82

imer

0.77

warm

0.77

rette

0.76

ounge

0.74

ansk

0.74

udic

0.73

ayette

0.70

Activations Density 0.562%

proper nouns related to various people or entities

names or terms related to specific entities or organizations

No Comments

No Known Activations