INDEX

Explanations

descriptions and evaluations of various products and their effects

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

uum

-0.76

iates

-0.72

iate

-0.70

urches

-0.69

 defeats

-0.67

perate

-0.66

 harms

-0.63

powers

-0.62

ends

-0.62

ples

-0.62

POSITIVE LOGITS

 undoubtedly

1.07

ometric

0.87

 definitely

0.85

ovie

0.84

 probably

0.83

 indeed

0.82

 doubtless

0.80

 called

0.78

 Solitaire

0.78

 termed

0.75

Activations Density 0.162%

descriptions and evaluations of various products and their effects

No Comments

No Known Activations