INDEX

Explanations

terms or phrases that have been coined or used in a specific context

oai_token-act-pair · gpt-3.5-turbo

specific terminology and phrases that refer to concepts and labels

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Guerrero

-0.67

 choir

-0.67

 Bagg

-0.67

âĹ¼

-0.62

ierrez

-0.62

artney

-0.61

Reply

-0.61

 balcon

-0.61

 Brah

-0.59

isner

-0.58

POSITIVE LOGITS

 coined

1.05

icide

0.92

ifier

0.88

ounced

0.85

ename

0.84

 interchange

0.81

lance

0.81

assian

0.80

aran

0.80

ology

0.79

Activations Density 0.037%

terms or phrases that have been coined or used in a specific context

specific terminology and phrases that refer to concepts and labels

No Comments

No Known Activations