INDEX

Explanations

information related to media or media content

oai_token-act-pair · gpt-3.5-turbo

references to media content

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

SOM

-0.68

kun

-0.67

abouts

-0.67

 salsa

-0.65

0000000000000000

-0.65

Osc

-0.65

kson

-0.62

cake

-0.62

akeru

-0.62

igraph

-0.61

POSITIVE LOGITS

 playback

1.21

 outlets

0.91

 Coverage

0.91

eval

0.86

 inquiries

0.83

 Release

0.79

 outlet

0.78

 enqu

0.77

eals

0.76

 Matters

0.76

Activations Density 0.019%

information related to media or media content

references to media content

No Comments

No Known Activations