INDEX

Explanations

names of political leaders and figures

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

AMS

-0.71

ãĤ¤ãĥĪ

-0.71

ams

-0.71

bilt

-0.70

earch

-0.70

ãģĨ

-0.69

 Spielberg

-0.69

ãĥĵ

-0.68

ASE

-0.68

xp

-0.66

POSITIVE LOGITS

 Maduro

1.05

ascus

0.91

 Jinping

0.84

 regime

0.75

 dictatorship

0.73

 overth

0.72

 dictator

0.71

orsi

0.70

ensis

0.69

 denounced

0.69

Activations Density 0.006%

names of political leaders and figures

No Comments

No Known Activations