INDEX

Explanations

scenarios or situations described in a narrative form

oai_token-act-pair · gpt-3.5-turbo

mentions of various scenarios within discussions

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

ighters

-0.95

vals

-0.85

emouth

-0.84

omer

-0.83

ighter

-0.83

alties

-0.83

inking

-0.81

itsch

-0.80

oves

-0.78

interest

-0.77

POSITIVE LOGITS

 scenario

1.30

 scenarios

1.21

enario

0.85

eers

0.85

0.84

0.82

0.80

 shenan

0.80

-+-+

0.77

0.76

Activations Density 0.007%

scenarios or situations described in a narrative form

mentions of various scenarios within discussions

No Comments

No Known Activations