INDEX

Explanations

phrases related to popularity

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

uran

-0.76

erm

-0.72

INAL

-0.71

 Shell

-0.69

endez

-0.67

 Dull

-0.64

aer

-0.64

inis

-0.63

 Neurolog

-0.62

 Fury

-0.59

POSITIVE LOGITS

ability

0.95

Reviewer

0.87

ately

0.86

 ratings

0.85

quo

0.78

itism

0.77

ously

0.77

 tremend

0.75

rating

0.75

rise

0.75

Activations Density 0.046%

phrases related to popularity

No Comments

No Known Activations