INDEX

Explanations

references to upcoming events or future occurrences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

torch.float32

Hook Point

blocks.8.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

baugh

-0.74

 Feldman

-0.74

ocker

-0.72

ordes

-0.71

enance

-0.70

hess

-0.62

lees

-0.62

acers

-0.61

cks

-0.60

ans

-0.60

POSITIVE LOGITS

 week

1.16

 month

1.10

 year

1.00

 semester

0.95

ĻĤ

0.85

 door

0.83

door

0.81

decade

0.81

week

0.81

 Month

0.80

Activations Density 0.024%

references to upcoming events or future occurrences

No Comments

No Known Activations