INDEX

Explanations

references to escalating conflicts or disputes

oai_token-act-pair · gpt-3.5-turbo

references to escalating tensions in various contexts

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

Interstitial

-0.83

cise

-0.83

prints

-0.76

cause

-0.73

title

-0.73

icle

-0.73

sites

-0.70

icles

-0.69

missions

-0.69

AMA

-0.69

POSITIVE LOGITS

 tensions

1.14

 tension

0.98

 flared

0.92

 guiActiveUn

0.88

 arising

0.83

 escalation

0.81

 cooker

0.80

 escalate

0.78

 stemming

0.78

 provocation

0.77

Activations Density 0.015%

references to escalating conflicts or disputes

references to escalating tensions in various contexts

No Comments

No Known Activations