INDEX

Explanations

phrases related to auditory perception or information heard

oai_token-act-pair · gpt-3.5-turbo

instances of the word "heard."

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

redo

-0.85

prime

-0.76

eros

-0.75

aging

-0.74

esa

-0.73

sis

-0.73

efficient

-0.72

idad

-0.72

inals

-0.72

tan

-0.71

POSITIVE LOGITS

 heard

1.11

 Heard

1.11

 Hear

1.02

heard

0.96

 hears

0.93

 hear

0.92

 hearing

0.92

 Hearing

0.89

 voices

0.89

 listened

0.84

Activations Density 0.016%

phrases related to auditory perception or information heard

instances of the word "heard."

No Comments

No Known Activations