INDEX

Explanations

phrases related to locations and organizations

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

ãĤ´ãĥ³

-0.74

leon

-0.72

matter

-0.71

abre

-0.68

inventoryQuantity

-0.68

-0.63

depending

-0.63

fly

-0.63

azing

-0.61

romy

-0.61

POSITIVE LOGITS

1.46

("

1.36

('

1.16

([

1.09

(?,

1.08

(.

1.03

[(

1.00

($

0.99

()

0.98

0.97

Activations Density 1.073%

phrases related to locations and organizations

No Comments

No Known Activations