INDEX

Explanations

keywords related to health, research, and studies

oai_token-act-pair · gpt-3.5-turbo

sentences related to dietary habits and health metrics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

rament

-0.88

 bonded

-0.80

 facult

-0.78

 rede

-0.77

 diplomacy

-0.73

hug

-0.72

 servicing

-0.72

 defiant

-0.71

 unden

-0.70

 pardon

-0.70

POSITIVE LOGITS

RESULTS

1.52

 Results

1.40

 Participants

1.34

 Researchers

1.31

Results

1.23

 Using

1.20

 Each

1.17

 Subjects

1.16

 Turns

1.14

results

1.12

Activations Density 0.252%

keywords related to health, research, and studies

sentences related to dietary habits and health metrics

No Comments

No Known Activations