INDEX

Explanations

mentions of percentage values and numerical data

oai_token-act-pair · gpt-3.5-turbo

statistical measures and percentages related to healthcare or prescriptions

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

????????

-0.65

 propri

-0.60

 delicate

-0.59

 intric

-0.58

sic

-0.56

 incarn

-0.54

sonian

-0.54

fac

-0.54

 eccentric

-0.54

 someday

-0.54

POSITIVE LOGITS

 Median

1.22

 median

1.18

GDP

1.09

 Percentage

1.05

 Unemployment

1.02

 Gallup

1.01

 respondents

1.00

 unemployment

0.96

(%)

0.95

 households

0.94

Activations Density 0.855%

mentions of percentage values and numerical data

statistical measures and percentages related to healthcare or prescriptions

No Comments

No Known Activations