INDEX

Explanations

terms related to computer programming and software development, especially focusing on elements like code snippets 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_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

 charism

-0.53

 Lieberman

-0.49

 millenn

-0.49

 archived

-0.48

 Carney

-0.47

 Hollande

-0.46

 Indra

-0.46

 bolstered

-0.44

 supplemented

-0.44

 Ezek

-0.43

POSITIVE LOGITS

Os

0.54

omorphic

0.53

omorph

0.52

().

0.52

ometric

0.50

coin

0.49

udic

0.49

ratom

0.48

otype

0.48

pi

0.48

Activations Density 6.460%

terms related to computer programming and software development, especially focusing on elements like code snippets and technical instructions

No Comments

No Known Activations