INDEX

Explanations

phrases containing non-English characters and a mix of English words

oai_token-act-pair · gpt-3.5-turbo

specific characters or symbols that may indicate encoded or corrupted text

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

boys

-0.63

accompan

-0.61

 bearer

-0.60

packing

-0.59

multipl

-0.59

theless

-0.58

 overload

-0.58

DragonMagazine

-0.55

 besides

-0.55

watching

-0.54

POSITIVE LOGITS

´

1.41

¤

1.41

¶

1.39

²

1.36

¾

1.34

¬

1.30

¦

1.29

ĥ

1.27

¼

1.27

«

1.26

Activations Density 0.074%

phrases containing non-English characters and a mix of English words

specific characters or symbols that may indicate encoded or corrupted text

No Comments

No Known Activations