INDEX

Explanations

terms related to the intensification or increase of a situation, specifically focusing on potential conflicts or tensions

oai_token-act-pair · gpt-3.5-turbo

terms related to escalation in various contexts, particularly in conflict or tension scenarios

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Faith

-0.76

 Fortune

-0.72

raft

-0.67

Bra

-0.67

 Flor

-0.67

athe

-0.67

wig

-0.66

 Latter

-0.65

urn

-0.65

rie

-0.64

POSITIVE LOGITS

 escalate

3.27

 escalation

3.23

 escalated

2.93

 escalating

2.72

escal

2.54

 escal

2.33

 intens

1.40

 intensified

1.38

 intensify

1.33

 worsening

1.22

Activations Density 0.019%

terms related to the intensification or increase of a situation, specifically focusing on potential conflicts or tensions

terms related to escalation in various contexts, particularly in conflict or tension scenarios

No Comments

No Known Activations