INDEX

Explanations

mentions of specific symbols or characters

oai_token-act-pair · gpt-3.5-turbo

specific symbols or characters that may indicate special formatting or emotional emphasis in text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 unidentified

-0.64

ATM

-0.63

 mount

-0.63

Aur

-0.62

 intens

-0.62

 sund

-0.62

 manager

-0.61

 mont

-0.61

 range

-0.60

olar

-0.60

POSITIVE LOGITS

Ĺ

4.59

ĸ

2.10

ķ

1.82

ĺ

1.81

ļ

1.74

ĥ

1.73

¤

1.72

¦

1.72

ĵ

1.70

§

1.66

Activations Density 0.004%

mentions of specific symbols or characters

specific symbols or characters that may indicate special formatting or emotional emphasis in text

No Comments

No Known Activations