INDEX

Explanations

the word "id" followed by a numeric value

oai_token-act-pair · gpt-3.5-turbo

references to the term "id" and related concepts

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

vernment

-0.73

raise

-0.71

haps

-0.65

olitan

-0.62

gdala

-0.62

xual

-0.61

restrial

-0.61

gans

-0.61

 âĹı

-0.61

andel

-0.60

POSITIVE LOGITS

id

3.69

Id

1.91

Id

1.79

 idle

1.58

ID

1.51

id

1.35

pid

1.33

IDs

1.22

 Idle

1.19

idi

1.15

Activations Density 0.022%

the word "id" followed by a numeric value

references to the term "id" and related concepts

No Comments

No Known Activations