INDEX

Explanations

references to significant individuals or names associated with authority and influence

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

 fundamentals

-0.65

 bottleneck

-0.65

 execution

-0.64

 executing

-0.63

fet

-0.63

 stewards

-0.63

 diction

-0.62

 hypers

-0.61

 resil

-0.60

 organ

-0.60

POSITIVE LOGITS

)(

0.96

0.91

esides

0.89

MRI

0.87

ought

0.85

),"

0.84

undreds

0.84

oused

0.83

idav

0.82

ospons

0.82

Activations Density 0.048%

references to significant individuals or names associated with authority and influence

No Comments

No Known Activations