INDEX

Explanations

percentages and numerical figures

oai_token-act-pair · gpt-3.5-turbo

quantitative data and statistics related to percentages

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ãĥ¤

-0.75

 obser

-0.74

ITED

-0.73

pload

-0.72

 paran

-0.70

zek

-0.66

 behav

-0.65

 detrim

-0.65

åİ

-0.64

=-=-=-=-=-=-=-=-

-0.64

POSITIVE LOGITS

cake

0.87

percent

0.80

imet

0.78

rowth

0.78

iles

0.77

 percent

0.73

%%

0.73

rate

0.71

 Percent

0.66

cision

0.65

Activations Density 0.053%

percentages and numerical figures

quantitative data and statistics related to percentages

No Comments

No Known Activations