INDEX

Explanations

phrases related to disbelief or contradiction

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_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ourses

-0.82

ufact

-0.69

onential

-0.68

chairs

-0.67

lling

-0.67

erella

-0.67

rams

-0.67

-0.66

Course

-0.63

lled

-0.63

POSITIVE LOGITS

,.

0.83

0.79

 Â¯

0.73

 there

0.71

0.64

,,,,

0.64

,...

0.62

 congratulations

0.62

,,

0.61

âĶĢ

0.59

Activations Density 0.105%

phrases related to disbelief or contradiction

No Comments

No Known Activations