INDEX

Explanations

numerical figures related to quantities and statistics

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

↵Âł

-0.64

 wisely

-0.63

Helper

-0.62

-0.60

 Casey

-0.59

 Bomber

-0.57

ODE

-0.57

Johnny

-0.55

rade

-0.54

 elig

-0.54

POSITIVE LOGITS

 dozen

1.42

dozen

1.39

 hundred

1.22

 thousand

1.14

 million

1.04

 lakh

1.00

ousand

0.99

 billion

0.90

 thousands

0.90

 thirds

0.88

Activations Density 1.644%

numerical figures related to quantities and statistics

No Comments

No Known Activations