INDEX

Explanations

phrases related to controversial social, political, and ethical issues

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_fs6144-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_6144

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

6,144

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

aukee

-0.80

mouth

-0.73

Laughs

-0.71

Guard

-0.70

stal

-0.70

AL

-0.67

backer

-0.66

ETA

-0.66

wn

-0.65

Tank

-0.65

POSITIVE LOGITS

 someday

0.82

 whoever

0.76

 they

0.75

 justifies

0.73

 these

0.70

 although

0.69

 there

0.68

 policymakers

0.68

 such

0.67

 humans

0.65

Activations Density 0.396%

phrases related to controversial social, political, and ethical issues

No Comments

No Known Activations