INDEX

Explanations

words related to coding, programming, and software development processes, such as specific languages, code-related terms, and tools commonly used in the software industry

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

iatus

-0.84

 Flavoring

-0.80

esville

-0.79

abouts

-0.79

iscopal

-0.78

 Wynne

-0.72

ESA

-0.71

Ń·

-0.67

innie

-0.67

strous

-0.66

POSITIVE LOGITS

 snippet

1.34

 snippets

1.16

base

0.97

velop

0.93

 code

0.91

 snipp

0.89

code

0.88

 coded

0.83

 fragment

0.81

 interpreter

0.81

Activations Density 0.671%

words related to coding, programming, and software development processes, such as specific languages, code-related terms, and tools commonly used in the software industry

No Comments

No Known Activations