INDEX

Explanations

proper nouns or names

oai_token-act-pair · gpt-3.5-turbo

the presence of specific Pokémon character descriptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 thereafter

-0.66

 afterward

-0.65

 UNCLASSIFIED

-0.64

etheless

-0.61

enegger

-0.61

[*

-0.60

}.

-0.59

kson

-0.59

 afterwards

-0.59

_.

-0.59

POSITIVE LOGITS

 Brewing

0.63

reetings

0.61

Coin

0.61

coin

0.60

ratom

0.58

 âĢº

0.58

 Klux

0.56

][

0.56

oÄŁan

0.55

 Album

0.55

Activations Density 0.810%

proper nouns or names

the presence of specific Pokémon character descriptions

No Comments

No Known Activations