INDEX

Explanations

definitions, explanations, and key terms

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

©¶æ

-0.66

tsy

-0.63

otin

-0.60

oqu

-0.58

isol

-0.57

uten

-0.57

otos

-0.56

 suicides

-0.56

ython

-0.56

philis

-0.55

POSITIVE LOGITS

 Anyway

0.81

 compared

0.80

 versus

0.79

?,

0.77

.</

0.76

!?

0.70

 beforehand

0.69

Bottom

0.69

!.

0.68

!:

0.67

Activations Density 1.144%

definitions, explanations, and key terms

No Comments

No Known Activations