INDEX

Explanations

statements speculating about possibilities or potential scenarios

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

ilty

-0.70

plex

-0.69

ele

-0.64

 Need

-0.59

 Notting

-0.58

 Blitz

-0.58

gencies

-0.57

Lv

-0.57

bert

-0.56

utor

-0.56

POSITIVE LOGITS

 easily

1.02

 feas

0.98

ivably

0.93

 conce

0.92

ħĭ

0.90

 possibly

0.81

 potentially

0.78

heard

0.75

 disastrous

0.74

idon

0.74

Activations Density 0.106%

statements speculating about possibilities or potential scenarios

No Comments

No Known Activations