INDEX

Explanations

phrases indicating rarity and classification of items or entities

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

verning

-0.77

apo

-0.76

entric

-0.73

akespeare

-0.70

EStreamFrame

-0.69

CLS

-0.65

REP

-0.64

uckland

-0.64

illation

-0.63

 ensemble

-0.63

POSITIVE LOGITS

craft

0.88

hood

0.80

ore

0.80

 variants

0.80

 items

0.79

ãĥ¬

0.77

 metals

0.74

 coins

0.74

ness

0.73

coins

0.73

Activations Density 0.013%

phrases indicating rarity and classification of items or entities

No Comments

No Known Activations