INDEX

Explanations

proper nouns related to various entities or people

oai_token-act-pair · gpt-3.5-turbo

references to specific titles or names in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

soType

-0.75

xual

-0.74

abase

-0.71

azeera

-0.69

yout

-0.68

netflix

-0.68

wikipedia

-0.67

idia

-0.66

warm

-0.63

ucl

-0.61

POSITIVE LOGITS

 Furious

0.65

atoes

0.64

pload

0.64

ooters

0.63

unic

0.62

hesda

0.57

 Pepper

0.56

ibaba

0.56

monary

0.55

 tumble

0.55

Activations Density 0.348%

proper nouns related to various entities or people

references to specific titles or names in various contexts

No Comments

No Known Activations