INDEX

Explanations

terms related to aggressive behavior or actions

oai_token-act-pair · gpt-3.5-turbo

instances and discussions of aggression

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

FORMATION

-0.80

aver

-0.76

ectar

-0.72

Rated

-0.70

van

-0.67

rote

-0.66

 Vita

-0.66

âĢ¢âĢ¢

-0.65

 Mour

-0.65

clips

-0.64

POSITIVE LOGITS

 aggression

1.15

 aggress

0.93

iveness

0.86

 provocation

0.82

 towards

0.82

 toward

0.81

 aggressively

0.81

 escalation

0.79

Agg

0.79

 whine

0.75

Activations Density 0.011%

terms related to aggressive behavior or actions

instances and discussions of aggression

No Comments

No Known Activations