INDEX

Explanations

numbers or measurement-related terms

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

 Rica

-0.64

 Buster

-0.64

Shares

-0.63

 Clapper

-0.63

 Freed

-0.61

CHECK

-0.60

 cardinal

-0.58

igham

-0.56

cade

-0.56

 Calling

-0.56

POSITIVE LOGITS

distance

0.75

 intra

0.73

 carbohyd

0.66

entric

0.65

ishable

0.65

chard

0.64

wait

0.64

percent

0.64

orbit

0.63

 millennial

0.62

Activations Density 0.015%

numbers or measurement-related terms

No Comments

No Known Activations