INDEX

Explanations

phrases related to contribution or support

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

lication

-0.78

rants

-0.72

NetMessage

-0.71

letter

-0.66

llo

-0.65

anned

-0.64

ogl

-0.63

tun

-0.62

fing

-0.61

naire

-0.60

POSITIVE LOGITS

 shaping

0.77

 Flavoring

0.74

 improving

0.73

 society

0.69

 solving

0.68

 strengthening

0.67

 prosperity

0.66

 diversity

0.65

 strengthen

0.65

the

0.65

Activations Density 0.045%

phrases related to contribution or support

No Comments

No Known Activations