INDEX

Explanations

references to prominent figures and celebrities

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

igans

-0.69

soType

-0.65

helle

-0.63

 procedural

-0.63

earchers

-0.62

ufact

-0.62

externalActionCode

-0.61

leans

-0.60

Tile

-0.60

portation

-0.60

POSITIVE LOGITS

XVI

0.79

ogun

0.71

III

0.70

 Lenin

0.70

µ

0.67

çİĭ

0.67

äºº

0.65

vez

0.64

XIV

0.63

Ŀ

0.62

Activations Density 0.065%

references to prominent figures and celebrities

No Comments

No Known Activations