INDEX

Explanations

words related to labeling or categorizing, especially in the context of products or music releases

oai_token-act-pair · gpt-3.5-turbo

terms associated with music labels

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

acus

-0.67

================================================================

-0.65

olars

-0.65

otos

-0.64

alez

-0.63

thus

-0.62

umbnail

-0.62

Yin

-0.62

abama

-0.61

aptic

-0.61

POSITIVE LOGITS

 label

1.15

 labels

1.06

label

0.94

Label

0.90

 Label

0.88

red

0.74

Matrix

0.72

=""

0.72

cloth

0.72

liga

0.72

Activations Density 0.010%

words related to labeling or categorizing, especially in the context of products or music releases

terms associated with music labels

No Comments

No Known Activations