INDEX

Explanations

information related to product reviews, especially focused on ratings and feedback on the quality and performance of different products

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

onomous

-0.81

 linguistic

-0.80

Population

-0.78

ocracy

-0.76

 separatists

-0.75

 broadcasting

-0.75

 invoking

-0.74

 broadcaster

-0.73

 advers

-0.73

 ideological

-0.72

POSITIVE LOGITS

 Stainless

1.32

 silicone

1.25

 stainless

1.21

 packaging

1.20

 accessories

1.20

 adjustable

1.19

 matte

1.19

 waterproof

1.17

 removable

1.17

 Accessories

1.17

Activations Density 3.224%

information related to product reviews, especially focused on ratings and feedback on the quality and performance of different products

No Comments

No Known Activations