INDEX

Explanations

references to different types of games

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

ayers

-0.69

 acknow

-0.68

 politic

-0.67

å§«

-0.66

umper

-0.64

 unnamed

-0.63

ignt

-0.63

undai

-0.62

ricane

-0.62

thening

-0.62

POSITIVE LOGITS

wright

1.02

arium

0.97

PLAY

0.95

play

0.93

pad

0.92

cube

0.90

manship

0.90

eers

0.88

plays

0.82

eer

0.82

Activations Density 0.047%

references to different types of games

No Comments

No Known Activations