INDEX

Explanations

references to combat actions related to fighting at close range

oai_token-act-pair · gpt-3.5-turbo

terms related to melee combat and fighting styles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

-0.82

arte

-0.79

 Cosponsors

-0.77

é¾įå

-0.73

 Enlight

-0.72

odo

-0.72

Catalog

-0.71

worthiness

-0.69

 Amendments

-0.69

odor

-0.69

POSITIVE LOGITS

 combatants

0.97

 weapons

0.88

 melee

0.85

 wielding

0.84

 weaponry

0.84

 combat

0.80

halla

0.79

 attackers

0.78

 weapon

0.76

erness

0.74

Activations Density 0.018%

references to combat actions related to fighting at close range

terms related to melee combat and fighting styles

No Comments

No Known Activations