INDEX

Explanations

code-related keywords and phrases

oai_token-act-pair · gpt-3.5-turbo

references to code and programming-related concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Flavoring

-0.84

 Wynne

-0.79

iscopal

-0.76

ESA

-0.76

iatus

-0.72

abouts

-0.69

 Sachs

-0.68

iversal

-0.67

Ń·

-0.66

 Bever

-0.65

POSITIVE LOGITS

 snippet

1.41

 snippets

1.30

base

1.26

 snipp

0.98

 repository

0.93

 reuse

0.92

velop

0.91

code

0.88

 repositories

0.87

unit

0.86

Activations Density 0.024%

code-related keywords and phrases

references to code and programming-related concepts

No Comments

No Known Activations