INDEX

Explanations

sentence endings, particularly ones emphasizing finality or conclusions

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hee

-0.67

 acidic

-0.63

ub

-0.59

æ©Ł

-0.58

available

-0.57

esome

-0.56

kaya

-0.56

Dou

-0.56

pload

-0.55

RNA

-0.55

POSITIVE LOGITS

angering

1.41

owment

1.39

ocrine

1.13

ocrin

1.11

angers

1.09

eared

1.00

angered

1.00

orph

1.00

urance

0.96

orses

0.96

Activations Density 0.512%

sentence endings, particularly ones emphasizing finality or conclusions

No Comments

No Known Activations