INDEX

Explanations

statements implying a sense of realization or conclusion

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

aukee

-0.83

mouth

-0.78

andem

-0.76

aunder

-0.71

cellaneous

-0.68

adr

-0.68

aturally

-0.67

eta

-0.67

ivalent

-0.67

Ĥİ

-0.67

POSITIVE LOGITS

 there

0.90

 none

0.79

 whoever

0.78

 nobody

0.77

 although

0.74

 awaits

0.73

 they

0.73

 neither

0.73

we

0.68

 exists

0.67

Activations Density 0.188%

statements implying a sense of realization or conclusion

No Comments

No Known Activations