INDEX

Explanations

references to geographical 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

 UNCLASSIFIED

-0.85

 skelet

-0.71

etheless

-0.61

âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ

-0.60

âĹ¼

-0.60

iferation

-0.58

 exting

-0.58

 Azerb

-0.56

 disadvant

-0.55

pse

-0.54

POSITIVE LOGITS

 [âĢ¦]

0.68

 âĢİ

0.68

 âĢº

0.62

âĢİ

0.60

 Donald

0.56

ADVERTISEMENT

0.52

Welcome

0.52

 ....

0.51

↵Âł

0.50

 frontman

0.50

Activations Density 5.146%

references to geographical locations

No Comments

No Known Activations