INDEX

Explanations

phrases indicating repetition or frequency

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

HCR

-0.70

skirts

-0.69

Ü

-0.68

Ú

-0.66

FORE

-0.65

 Procedures

-0.65

 ********************************

-0.64

Ø±

-0.64

 streng

-0.64

arty

-0.63

POSITIVE LOGITS

 conceivable

1.00

 ounce

0.98

 dollar

0.97

 hundred

0.82

THING

0.81

 pound

0.81

 thousand

0.80

 succeeding

0.79

 imaginable

0.79

 successive

0.78

Activations Density 0.019%

phrases indicating repetition or frequency

No Comments

No Known Activations