INDEX

Explanations

capitalized words with 'Im' as a prefix

oai_token-act-pair · gpt-3.5-turbo

instances of the prefix "Im-" indicating immediate actions or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

NetMessage

-0.97

 stakes

-0.74

 grapes

-0.73

 shroud

-0.71

 Titanic

-0.71

ttes

-0.70

 wool

-0.69

 braces

-0.68

 Nazis

-0.66

 asbestos

-0.66

POSITIVE LOGITS

perial

1.47

mediately

1.46

mediate

1.43

plementation

1.40

plement

1.37

ple

1.37

ply

1.31

minent

1.28

posing

1.27

perfect

1.24

Activations Density 0.023%

capitalized words with 'Im' as a prefix

instances of the prefix "Im-" indicating immediate actions or conditions

No Comments

No Known Activations