INDEX

Explanations

phrases indicating a change or development in a positive direction

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

Joined

-0.74

odied

-0.74

 successors

-0.69

otype

-0.68

 predecessors

-0.68

DNA

-0.67

Introduced

-0.65

issance

-0.65

inia

-0.64

lication

-0.62

POSITIVE LOGITS

 downhill

0.90

 spir

0.79

 BELOW

0.77

 blurry

0.73

 unfolded

0.73

 murky

0.72

 bleak

0.70

 calmed

0.70

 fluid

0.70

 Thrones

0.70

Activations Density 1.657%

phrases indicating a change or development in a positive direction

No Comments

No Known Activations