INDEX

Explanations

words related to combustion and firing

oai_token-act-pair · gpt-3.5-turbo

terms related to different types of combustion and fire-associated events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Zone

-0.73

 Hanson

-0.65

 races

-0.65

 break

-0.63

 race

-0.63

Yen

-0.62

 Poly

-0.62

 superv

-0.61

 abstract

-0.61

sub

-0.61

POSITIVE LOGITS

fired

4.41

fires

1.82

burning

1.53

 Fired

1.40

fire

1.36

laun

1.12

fed

1.02

dated

1.02

coal

0.96

faced

0.95

Activations Density 0.013%

words related to combustion and firing

terms related to different types of combustion and fire-associated events

No Comments

No Known Activations