INDEX

Explanations

locations or actions involving aggression or conflict

oai_token-act-pair · gpt-3.5-turbo

instances of physical interactions or confrontations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Included

-0.62

 Laksh

-0.61

mine

-0.61

anish

-0.59

Quantity

-0.58

fried

-0.57

orno

-0.57

sonian

-0.57

FIG

-0.56

 Owner

-0.56

POSITIVE LOGITS

 whom

0.87

oba

0.71

armac

0.70

 waivers

0.67

acebook

0.65

ventory

0.64

insky

0.64

 stride

0.62

vl

0.62

 subtle

0.60

Activations Density 0.580%

locations or actions involving aggression or conflict

instances of physical interactions or confrontations

No Comments

No Known Activations