INDEX

Explanations

phrases surrounded by quotation marks indicating a direct quote

oai_token-act-pair · gpt-3.5-turbo

quoted speech or statements

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Ross

-0.77

rul

-0.74

 Versus

-0.74

 Battles

-0.73

 derby

-0.72

 Mehran

-0.72

XI

-0.71

 rall

-0.70

Wi

-0.70

 affiliate

-0.69

POSITIVE LOGITS

absolutely

1.75

completely

1.63

very

1.61

pretty

1.61

probably

1.61

extremely

1.59

almost

1.54

highly

1.52

quite

1.52

great

1.50

Activations Density 0.129%

phrases surrounded by quotation marks indicating a direct quote

quoted speech or statements

No Comments

No Known Activations