INDEX

Explanations

phrases related to conflict or competition

oai_token-act-pair · gpt-3.5-turbo

words related to being trapped or constrained

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

 Clinic

-0.63

 paraly

-0.62

 Patterson

-0.59

 antit

-0.58

Subscribe

-0.56

OUN

-0.54

 extent

-0.53

KN

-0.53

iod

-0.53

annabin

-0.52

POSITIVE LOGITS

ged

4.57

ging

2.99

ges

2.36

gers

2.02

ger

1.93

ge

1.82

gered

1.77

gement

1.76

gery

1.64

gence

1.62

Activations Density 0.008%

phrases related to conflict or competition

words related to being trapped or constrained

No Comments

No Known Activations