INDEX

Explanations

instances where the word "listen" is mentioned

oai_token-act-pair · gpt-3.5-turbo

instances of the word "listen" related to audio consumption or attention

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ãĤ¨ãĥ«

-0.66

 Athletics

-0.61

VAT

-0.60

 Staples

-0.60

enture

-0.60

cil

-0.60

Lago

-0.59

eros

-0.59

grave

-0.59

 Glory

-0.58

POSITIVE LOGITS

 listen

1.43

 listened

1.36

 listens

1.33

 listening

1.21

 Listen

1.12

 attent

1.03

Listen

0.99

 listener

0.95

tracks

0.92

 listeners

0.88

Activations Density 0.010%

instances where the word "listen" is mentioned

instances of the word "listen" related to audio consumption or attention

No Comments

No Known Activations