INDEX

Explanations

words related to technology and technical instructions

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_fs768-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

768

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

 Lancaster

-0.62

 Samar

-0.61

 notebooks

-0.61

 Skydragon

-0.61

 Tasman

-0.61

 Movies

-0.60

 Stuff

-0.60

 partnerships

-0.60

Yor

-0.60

Sud

-0.59

POSITIVE LOGITS

ï¸ı

1.02

agree

0.97

vest

0.96

own

0.95

ve

0.93

tu

0.89

reci

0.87

cre

0.86

ved

0.85

Activations Density 1.866%

words related to technology and technical instructions

No Comments

No Known Activations