INDEX

Explanations

direct references to cause-and-effect relationships or consequences

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

gerald

-0.77

anners

-0.75

glers

-0.71

 Frie

-0.65

 cautiously

-0.64

 mell

-0.63

abies

-0.62

 Chocobo

-0.62

Sleep

-0.62

 safely

-0.60

POSITIVE LOGITS

forward

0.84

ebted

0.81

 contradicted

0.80

 contradicts

0.73

 sunlight

0.71

 observable

0.71

orship

0.70

oire

0.69

achable

0.68

aneous

0.68

Activations Density 0.534%

direct references to cause-and-effect relationships or consequences

No Comments

No Known Activations