INDEX

Explanations

technical terms related to technology and technical documentation

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

 extract

-0.57

 Barg

-0.57

 confir

-0.57

 Levi

-0.57

 extracted

-0.54

-0.53

 ende

-0.53

 carriers

-0.52

alist

-0.52

 accounted

-0.52

POSITIVE LOGITS

0.91

,-

0.87

KEN

0.84

)=(

0.74

PIN

0.73

Merit

0.71

maxwell

0.69

ken

0.68

ypes

0.68

nai

0.68

Activations Density 1.045%

technical terms related to technology and technical documentation

No Comments

No Known Activations