INDEX

Explanations

the word "zone" at different levels of activation

oai_token-act-pair · gpt-3.5-turbo

references to different types of zones

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

sonian

-0.84

aldi

-0.83

anooga

-0.77

utive

-0.75

HCR

-0.73

thora

-0.72

tons

-0.71

akuya

-0.70

verified

-0.70

issance

-0.70

POSITIVE LOGITS

 zones

0.97

 zone

0.94

 dwell

0.91

 boundaries

0.75

 boundary

0.73

 deline

0.72

 Areas

0.72

 Zone

0.71

 crawling

0.70

 dotted

0.69

Activations Density 0.043%

the word "zone" at different levels of activation

references to different types of zones

No Comments

No Known Activations