INDEX

Explanations

proper nouns in the format "Di[...]" that seem to be related to individuals

oai_token-act-pair · gpt-3.5-turbo

names and identifiers related to individuals and entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ãĤ´ãĥ³

-0.68

ĵĺ

-0.59

DAQ

-0.58

 Jagu

-0.57

HI

-0.56

 heels

-0.55

 fries

-0.55

ignty

-0.54

 pains

-0.53

 glow

-0.53

POSITIVE LOGITS

andowski

0.81

pora

0.81

govtrack

0.75

weekly

0.71

cious

0.69

rences

0.67

anni

0.66

 Alley

0.66

ython

0.63

ento

0.62

Activations Density 0.165%

proper nouns in the format "Di[...]" that seem to be related to individuals

names and identifiers related to individuals and entities

No Comments

No Known Activations