INDEX

Explanations

terms related to programming functions, particularly in functional programming and computer science contexts

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

 entreprene

-0.68

indust

-0.66

naissance

-0.65

 Niet

-0.64

zona

-0.64

 invitations

-0.63

introdu

-0.63

WARE

-0.62

Ire

-0.61

-0.60

POSITIVE LOGITS

0.99

^{

0.89

=>

0.86

<=

0.83

->

0.83

 âĨĴ

0.83

++

0.82

 âī¥

0.81

0.79

Activations Density 0.026%

terms related to programming functions, particularly in functional programming and computer science contexts

No Comments

No Known Activations