INDEX

Explanations

terms related to cycling or changing states, specifically in the context of "revolving" or "rotating."

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

xual

-0.88

edly

-0.85

anas

-0.83

essors

-0.80

anan

-0.79

Phys

-0.77

hammad

-0.76

riel

-0.76

00200000

-0.75

imates

-0.75

POSITIVE LOGITS

 revolving

0.99

 door

0.93

 staircase

0.85

 wheel

0.83

 circle

0.83

 bolt

0.79

 stair

0.76

 rotating

0.72

 drum

0.72

 Door

0.72

Activations Density 0.009%

terms related to cycling or changing states, specifically in the context of "revolving" or "rotating."

No Comments

No Known Activations