INDEX

Explanations

proper nouns or names

oai_token-act-pair · gpt-3.5-turbo

proper names, particularly those of individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

psons

-0.81

culosis

-0.80

otype

-0.74

isation

-0.73

arial

-0.71

iator

-0.70

early

-0.70

ization

-0.69

 wagon

-0.68

etheless

-0.68

POSITIVE LOGITS

 Bass

0.88

 Shack

0.85

Hug

0.82

Poll

0.81

 Guard

0.81

De

0.80

 Mash

0.79

Bot

0.78

Rum

0.77

Disc

0.77

Activations Density 0.013%

proper nouns or names

proper names, particularly those of individuals

No Comments

No Known Activations