INDEX

Explanations

numbers indicating distances in miles

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

eer

-0.83

gotten

-0.70

icer

-0.70

gio

-0.70

olith

-0.69

icious

-0.68

izers

-0.67

enum

-0.67

eers

-0.64

 inval

-0.63

POSITIVE LOGITS

 radius

0.95

 traveled

0.86

 kilometers

0.86

km

0.83

km

0.81

 kilometres

0.81

 distance

0.79

 inland

0.77

 travelled

0.77

 miles

0.77

Activations Density 0.031%

numbers indicating distances in miles

No Comments

No Known Activations