INDEX

Explanations

elements related to mathematical notation, specifically those pertaining to sets and cardinality

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

 spills

-0.64

 clinch

-0.62

 overdue

-0.60

 seiz

-0.60

gered

-0.60

 rolls

-0.60

jog

-0.59

 Pigs

-0.59

 conduc

-0.58

 sabotage

-0.58

POSITIVE LOGITS

{\

1.08

\)

0.99

0.98

0.96

_{

0.94

}\

0.93

actionDate

0.91

²

0.86

}}

0.85

{\

0.84

Activations Density 0.035%

elements related to mathematical notation, specifically those pertaining to sets and cardinality

No Comments

No Known Activations