INDEX

Explanations

mentions of specific names like "Kaz" and "Haj" in the text

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names related to individuals and places

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

 Pell

-0.73

irm

-0.68

PE

-0.67

cli

-0.63

ede

-0.62

payer

-0.62

ronic

-0.62

 provider

-0.62

rained

-0.61

 institutional

-0.61

POSITIVE LOGITS

Kaz

3.97

Koz

1.77

 Kats

1.57

 Kazakhstan

1.51

Haj

1.47

Kas

1.46

 Kuro

1.42

 Kazakh

1.41

Kab

1.38

 Kash

1.37

Activations Density 0.024%

mentions of specific names like "Kaz" and "Haj" in the text

proper nouns, particularly names related to individuals and places

No Comments

No Known Activations