INDEX

Explanations

details or information related to investigations, research, or reports

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_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

holders

-0.83

ngth

-0.75

 Reincarn

-0.73

 bearer

-0.70

 subsistence

-0.70

astern

-0.70

anguage

-0.66

 thirds

-0.65

 combust

-0.65

mount

-0.65

POSITIVE LOGITS

ienne

1.31

iana

1.22

iane

1.18

iets

1.15

ian

1.15

iano

1.14

iani

1.11

ians

1.11

iom

1.10

iet

1.08

Activations Density 0.026%

details or information related to investigations, research, or reports

No Comments

No Known Activations