INDEX

Explanations

phrases related to interventions or decision-making

oai_token-act-pair · gpt-3.5-turbo

instances of special characters or symbols in the text

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

 Brist

-0.73

enegger

-0.72

 Thomson

-0.70

 airs

-0.67

itsch

-0.65

 Manhattan

-0.65

Tob

-0.64

 Strat

-0.64

 Shap

-0.63

 Borough

-0.63

POSITIVE LOGITS

¬

1.60

Ļ

1.49

ı

1.23

ĸ

1.21

ª

1.17

¾

1.16

ļ

1.15

¡

1.15

ħ

1.14

ľ

1.13

Activations Density 0.384%

phrases related to interventions or decision-making

instances of special characters or symbols in the text

No Comments

No Known Activations