INDEX

Explanations

phrases emphasizing the exclusivity or uniqueness of something

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

idon

-0.76

insula

-0.71

ahead

-0.64

align

-0.64

comb

-0.59

ducers

-0.57

phen

-0.57

dict

-0.57

multipl

-0.56

finder

-0.54

POSITIVE LOGITS

 marginally

1.08

 kidding

0.81

ices

0.81

 spor

0.79

ICES

0.78

 incidentally

0.77

 lasted

0.75

 lasts

0.75

 scratches

0.71

 scratched

0.70

Activations Density 0.886%

phrases emphasizing the exclusivity or uniqueness of something

No Comments

No Known Activations