INDEX

Explanations

references to specific file formats and technical details

oai_token-act-pair · gpt-3.5-turbo

references to detailed reports or documentation

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 curfew

-0.77

 destiny

-0.70

 awoken

-0.69

 Reincarnated

-0.69

 racer

-0.67

 Lover

-0.67

 Warfare

-0.66

 Dress

-0.66

 Elys

-0.65

intendo

-0.61

POSITIVE LOGITS

 paragraphs

1.45

 pages

1.35

 excerpts

1.34

 concise

1.32

PDF

1.31

 excerpt

1.29

 readable

1.28

pdf

1.23

 explanatory

1.22

pdf

1.22

Activations Density 0.486%

references to specific file formats and technical details

references to detailed reports or documentation

No Comments

No Known Activations