INDEX

Explanations

references to programs or initiatives

oai_token-act-pair · gpt-3.5-turbo

references to various programs

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

pick

-0.69

 tales

-0.68

ModLoader

-0.66

 dilig

-0.64

 Robots

-0.64

uple

-0.63

SO

-0.62

âĶĢâĶĢâĶĢâĶĢ

-0.61

 IMAGES

-0.61

picking

-0.60

POSITIVE LOGITS

mable

2.09

matic

1.73

matically

1.70

mes

1.19

mers

1.15

matical

0.98

 Gutenberg

0.97

ming

0.96

atically

0.93

manship

0.88

Activations Density 0.040%

references to programs or initiatives

references to various programs

No Comments

No Known Activations