INDEX

Explanations

responses or reactions to various stimuli or events

oai_token-act-pair · gpt-3.5-turbo

phrases related to responding or reactions to events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

âĸ¬âĸ¬

-0.79

BALL

-0.76

gling

-0.74

 devils

-0.72

66666666

-0.69

 merger

-0.65

 whispers

-0.65

 Galileo

-0.65

Ò

-0.65

âĸ¬

-0.63

POSITIVE LOGITS

onds

1.27

onding

1.22

onder

1.18

awn

1.15

ond

1.00

endant

0.97

endants

0.92

uted

0.92

inished

0.90

ublic

0.90

Activations Density 0.016%

responses or reactions to various stimuli or events

phrases related to responding or reactions to events

No Comments

No Known Activations