INDEX

Explanations

phrases related to escalation or exacerbation of situations

oai_token-act-pair · gpt-3.5-turbo

terms related to rising tensions or conflicts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

prints

-0.85

lean

-0.83

leans

-0.78

print

-0.78

lication

-0.73

ribution

-0.72

ufact

-0.72

ciples

-0.71

inates

-0.70

itte

-0.70

POSITIVE LOGITS

 tensions

1.35

 tension

1.12

 hostilities

1.09

 conflict

1.00

 escalate

0.97

 escalation

0.94

 aggrav

0.92

 misunderstand

0.92

 worsening

0.91

 escalating

0.91

Activations Density 0.074%

phrases related to escalation or exacerbation of situations

terms related to rising tensions or conflicts

No Comments

No Known Activations