INDEX

Explanations

proper nouns, particularly names

oai_token-act-pair · gpt-3.5-turbo

instances of the word "na" 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 @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 desk

-0.73

 Martian

-0.66

 figure

-0.65

 Sounders

-0.62

 interstellar

-0.60

 heck

-0.59

 force

-0.59

 assembled

-0.59

 JPEG

-0.57

 bounce

-0.57

POSITIVE LOGITS

na

4.43

nas

2.13

nah

1.91

ni

1.81

ne

1.62

NA

1.61

nan

1.58

nes

1.56

ny

1.56

nia

1.48

Activations Density 0.010%

proper nouns, particularly names

instances of the word "na" in various contexts

No Comments

No Known Activations