INDEX

Explanations

past participles and forms of the verb "to be" indicating involvement

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

elling

-0.65

ovi

-0.64

enberg

-0.64

numbered

-0.63

Ĥİ

-0.63

ÃĥÃĤ

-0.62

 favour

-0.61

 partName

-0.61

published

-0.60

generation

-0.60

POSITIVE LOGITS

ment

0.85

igate

0.81

eering

0.77

 therein

0.74

igation

0.73

igating

0.73

 dots

0.72

rained

0.72

ttes

0.71

aneously

0.71

Activations Density 0.063%

past participles and forms of the verb "to be" indicating involvement

No Comments

No Known Activations