INDEX

Explanations

phrases related to consequential actions or events

oai_token-act-pair · gpt-3.5-turbo

instances of the conjunction "and" that link various ideas or concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

:(

-0.62

:[

-0.57

uta

-0.56

Ĥİ

-0.56

Â¶

-0.51

Shut

-0.50

)[

-0.50

.<

-0.49

Deal

-0.49

wire

-0.48

POSITIVE LOGITS

 consequently

0.93

 hence

0.91

 therefore

0.90

 thus

0.89

 thereby

0.83

 then

0.80

 vice

0.74

 secondly

0.72

 possibly

0.72

 thence

0.72

Activations Density 0.787%

phrases related to consequential actions or events

instances of the conjunction "and" that link various ideas or concepts

No Comments

No Known Activations