INDEX

Explanations

questions or statements expressing confusion or uncertainty

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

aughed

-0.88

lator

-0.84

ibaba

-0.71

opic

-0.70

haul

-0.66

iece

-0.66

chuk

-0.65

iHUD

-0.62

arnaev

-0.62

ãĤ¹

-0.61

POSITIVE LOGITS

soever

0.87

 someone

0.79

 somebody

0.79

 exactly

0.77

 bother

0.75

 anyone

0.72

why

0.72

 anybody

0.71

 people

0.68

 they

0.68

Activations Density 0.026%

questions or statements expressing confusion or uncertainty

No Comments

No Known Activations