INDEX

Explanations

phrases related to a list of items or reasons

oai_token-act-pair · gpt-3.5-turbo

instances of the word "including."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

orem

-0.79

inoa

-0.74

ļéĨĴ

-0.74

¬¼

-0.74

amus

-0.72

ulla

-0.72

ãĤº

-0.71

ablishment

-0.70

atorium

-0.69

pron

-0.68

POSITIVE LOGITS

 including

1.91

 namely

1.74

 such

1.65

including

1.57

 notably

1.55

such

1.51

 ranging

1.47

 includ

1.43

 Including

1.39

 particularly

1.12

Activations Density 0.502%

phrases related to a list of items or reasons

instances of the word "including."

No Comments

No Known Activations