INDEX

Explanations

changes or modifications made to something

oai_token-act-pair · gpt-3.5-turbo

terms related to modifications and updates

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

Fit

-0.60

 Fein

-0.57

 Knot

-0.56

 Cure

-0.56

rics

-0.55

dan

-0.53

eers

-0.53

 Nile

-0.53

lore

-0.51

AIDS

-0.51

POSITIVE LOGITS

 drastically

1.04

 accordingly

1.02

 dramatically

1.00

 substantially

0.97

 considerably

0.95

 significantly

0.95

 radically

0.89

 slightly

0.85

 greatly

0.84

 markedly

0.83

Activations Density 0.171%

changes or modifications made to something

terms related to modifications and updates

No Comments

No Known Activations