INDEX

Explanations

quantitative details and statistics

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_fs768-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

768

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

 Dynamics

-0.64

Express

-0.64

 derog

-0.63

ãĤ¿

-0.62

agate

-0.61

BT

-0.61

ffen

-0.59

NEY

-0.55

 cath

-0.54

inventoryQuantity

-0.53

POSITIVE LOGITS

 dozen

1.18

dozen

1.10

 eighty

0.96

thirds

0.95

 half

0.93

0.91

0.90

four

0.89

0.88

Activations Density 2.797%

quantitative details and statistics

No Comments

No Known Activations