INDEX

Explanations

measurements and dimensions related to physical objects

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

bara

-0.88

JECT

-0.73

gotten

-0.70

ctuary

-0.64

ibur

-0.64

uj

-0.63

roleum

-0.62

oras

-0.62

idem

-0.62

arded

-0.61

POSITIVE LOGITS

 ranges

0.96

 range

0.94

 estimation

0.86

 limit

0.85

 mismatch

0.85

 discrepancy

0.84

 difference

0.84

 measurement

0.83

 tolerance

0.80

 Limit

0.79

Activations Density 0.899%

measurements and dimensions related to physical objects

No Comments

No Known Activations

measurements and dimensions related to physical objects

No Comments

No Known Activations