INDEX

Explanations

references to intense standoffs or stalemates in different scenarios like conflicts, negotiations, or relationships

oai_token-act-pair · gpt-3.5-turbo

terms related to conflicts or confrontations

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

oline

-0.64

gae

-0.63

 Genius

-0.62

 samples

-0.60

iji

-0.60

 coloured

-0.60

Saharan

-0.60

 charities

-0.60

 Forms

-0.59

profits

-0.59

POSITIVE LOGITS

 standoff

3.62

 confrontation

1.94

 showdown

1.92

 shootout

1.78

 stal

1.58

 skirm

1.27

 Showdown

1.26

 clash

1.15

 brawl

1.14

 siege

1.12

Activations Density 0.030%

references to intense standoffs or stalemates in different scenarios like conflicts, negotiations, or relationships

terms related to conflicts or confrontations

No Comments

No Known Activations