INDEX

Explanations

references to specific "scenes" from movies or shows

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

orem

-0.92

achev

-0.82

rition

-0.79

dp

-0.79

reditary

-0.76

gements

-0.76

berman

-0.75

thood

-0.75

ikarp

-0.74

iott

-0.74

POSITIVE LOGITS

Dialogue

1.12

 scenes

0.99

 Scenes

0.98

 involving

0.97

 unfolding

0.92

Capture

0.91

 depicting

0.91

 Scene

0.86

 depicted

0.83

scene

0.82

Activations Density 0.046%

references to specific "scenes" from movies or shows

No Comments

No Known Activations