INDEX

Explanations

phrases indicating possession or ownership

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

istries

-0.77

WARE

-0.72

ãĥķãĤ©

-0.72

displayText

-0.69

nces

-0.69

paces

-0.68

 screenings

-0.68

bid

-0.67

hester

-0.66

ocations

-0.66

POSITIVE LOGITS

 unsuspecting

0.86

 deceased

0.75

 downed

0.68

dragon

0.65

 hydra

0.65

 giants

0.64

 aging

0.62

 deer

0.62

 cows

0.61

 fellow

0.61

Activations Density 0.068%

phrases indicating possession or ownership

No Comments

No Known Activations