INDEX

Explanations

terms related to scientific research, possibly focusing on particle physics and medical investigations

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 answ

-0.72

inarily

-0.69

azeera

-0.66

untled

-0.66

depending

-0.63

pload

-0.62

 lawy

-0.61

TPPStreamerBot

-0.61

dozen

-0.60

ãĤ´ãĥ³

-0.60

POSITIVE LOGITS

 Association

1.05

Ltd

0.90

 Edition

0.90

 Contest

0.88

LLP

0.86

 Competition

0.85

 Scholarship

0.82

LTD

0.81

naire

0.80

 âĵĺ

0.78

Activations Density 3.034%

terms related to scientific research, possibly focusing on particle physics and medical investigations

No Comments

No Known Activations