INDEX

Explanations

moments or instances of significance or importance

oai_token-act-pair · gpt-3.5-turbo

references to significant moments in narratives or experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

remlin

-0.76

GPU

-0.76

yu

-0.76

achev

-0.74

pel

-0.73

protein

-0.72

uv

-0.71

gon

-0.71

ply

-0.71

yss

-0.68

POSITIVE LOGITS

 moments

1.40

 Moments

1.19

 glimps

0.97

 moment

0.96

aneously

0.88

 Moment

0.87

 Minutes

0.84

 reminders

0.83

 gratification

0.82

 ILCS

0.77

Activations Density 0.011%

moments or instances of significance or importance

references to significant moments in narratives or experiences

No Comments

No Known Activations