INDEX

Explanations

references to specific objects or physical locations

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

idates

-0.82

 furthermore

-0.79

rity

-0.78

erity

-0.76

 moreover

-0.73

icals

-0.71

 additionally

-0.69

hess

-0.69

fficient

-0.67

iencies

-0.66

POSITIVE LOGITS

 proverbial

0.86

 sponge

0.86

 steroids

0.78

 miniature

0.77

 magnet

0.75

 Gest

0.74

 aspirin

0.73

yip

0.73

 heartbeat

0.72

 spaghetti

0.72

Activations Density 1.655%

references to specific objects or physical locations

No Comments

No Known Activations