INDEX

Explanations

information related to specific entities or technical details

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

tsy

-0.80

rette

-0.74

stad

-0.72

ohan

-0.68

sson

-0.68

AAAA

-0.67

ominated

-0.67

 Unsure

-0.66

roups

-0.66

hoe

-0.65

POSITIVE LOGITS

own

1.55

 predecessor

1.28

ELF

1.27

 namesake

1.24

 predecessors

1.20

 respective

1.06

 rightful

1.06

 constituent

1.03

 entirety

0.96

 infancy

0.95

Activations Density 0.632%

information related to specific entities or technical details

No Comments

No Known Activations