INDEX

Explanations

names and words with unusual characters

oai_token-act-pair · gpt-3.5-turbo

elements related to statistical or numerical references, particularly in the context of music or events

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

Hamilton

-1.08

Houston

-0.94

 Metatron

-0.89

 Harvey

-0.88

 Hamilton

-0.86

 Irving

-0.84

Texas

-0.80

 Holden

-0.78

CLASSIFIED

-0.78

Harris

-0.77

POSITIVE LOGITS

Äĩ

2.29

Äį

2.12

 Sloven

1.89

Å¡

1.86

Ä

1.74

Ä

1.72

 Croatian

1.70

 Serbia

1.68

 Serbian

1.66

 Croatia

1.64

Activations Density 0.195%

names and words with unusual characters

elements related to statistical or numerical references, particularly in the context of music or events

No Comments

No Known Activations