INDEX

Explanations

words related to listening or being attentive

oai_token-act-pair · gpt-3.5-turbo

instances of the word "listen" and its variations

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

eros

-0.68

Lago

-0.64

pex

-0.64

grave

-0.63

 Staples

-0.63

VAT

-0.62

 Athletics

-0.61

enture

-0.61

ball

-0.59

imgur

-0.59

POSITIVE LOGITS

 listened

1.20

 listen

1.20

 listens

1.19

 listening

1.14

 attent

1.10

 Listen

1.07

Listen

0.91

 Emin

0.89

 listener

0.84

 ears

0.82

Activations Density 0.013%

words related to listening or being attentive

instances of the word "listen" and its variations

No Comments

No Known Activations