INDEX

Explanations

terms related to physical and mechanical processes

oai_token-act-pair · gpt-3.5-turbo

connective phrases that imply relationships and conditions in explanations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

Justice

-0.82

Community

-0.81

aid

-0.76

asking

-0.75

polit

-0.74

Liberal

-0.73

ivals

-0.72

onomy

-0.72

moral

-0.72

OWN

-0.72

POSITIVE LOGITS

 protr

1.14

 moisture

1.05

 vibrations

1.05

 temperatures

1.04

 deposits

1.03

 airflow

1.03

 electrons

1.01

 gases

0.97

 fluids

0.96

 photons

0.95

Activations Density 0.382%

terms related to physical and mechanical processes

connective phrases that imply relationships and conditions in explanations

No Comments

No Known Activations