INDEX

Explanations

references to legal or procedural information

oai_token-act-pair · gpt-3.5-turbo

periods at the end of statements

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

ributes

-0.79

hug

-0.77

 quir

-0.75

 worsh

-0.75

 fishes

-0.73

 swim

-0.73

 cheering

-0.73

 portraits

-0.72

 favourite

-0.72

mur

-0.71

POSITIVE LOGITS

 Accordingly

1.50

 Therefore

1.47

 Additionally

1.46

 However

1.37

 Specifically

1.36

 Consequently

1.36

 Furthermore

1.36

 Moreover

1.33

 Thus

1.25

 Nevertheless

1.25

Activations Density 0.925%

references to legal or procedural information

periods at the end of statements

No Comments

No Known Activations