INDEX

Explanations

specific operation or mission names

oai_token-act-pair · gpt-3.5-turbo

references to military operations and their objectives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ervatives

-0.61

 hypocr

-0.59

agent

-0.59

 liar

-0.56

 Signs

-0.56

 hypocrisy

-0.55

*/(

-0.55

 bigot

-0.54

ocent

-0.54

notations

-0.53

POSITIVE LOGITS

 culminating

0.95

 culmin

0.94

 undertaken

0.87

 involving

0.86

yrinth

0.83

 culminated

0.82

Completed

0.79

 involves

0.79

pleted

0.78

 involve

0.78

Activations Density 1.191%

specific operation or mission names

references to military operations and their objectives

No Comments

No Known Activations