INDEX

Explanations

mentions of or references to the term "listener."

oai_token-act-pair · gpt-3.5-turbo

mentions of listeners and their interactions with content

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Athletics

-0.76

 Lancet

-0.64

 Carbuncle

-0.64

 Madrid

-0.62

 Leap

-0.62

 Sword

-0.62

 Ster

-0.62

 Rite

-0.62

 Fighting

-0.61

MSI

-0.61

POSITIVE LOGITS

 listener

1.67

 listeners

1.61

 listening

1.00

 listened

0.97

 tuned

0.94

 voic

0.92

iband

0.91

 answ

0.91

 listens

0.90

 ears

0.90

Activations Density 0.004%

mentions of or references to the term "listener."

mentions of listeners and their interactions with content

No Comments

No Known Activations