INDEX

Explanations

code-related terms and phrases

oai_token-act-pair · gpt-3.5-turbo

references to coding or programming concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 Flavoring

-0.95

 Wynne

-0.80

iatus

-0.77

iscopal

-0.76

ESA

-0.73

abouts

-0.73

iversal

-0.69

 Bever

-0.68

Ń·

-0.68

izo

-0.67

POSITIVE LOGITS

 snippet

1.35

 snippets

1.22

base

1.22

 snipp

0.96

code

0.92

 repository

0.92

velop

0.87

 fragment

0.85

 code

0.85

 reuse

0.81

Activations Density 0.029%

code-related terms and phrases

references to coding or programming concepts

No Comments

No Known Activations