INDEX

Explanations

verbs that describe negative political outcomes

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ç«

-0.68

 validity

-0.63

owe

-0.60

aminer

-0.58

 correlation

-0.58

orr

-0.58

 Haku

-0.57

utra

-0.57

 dated

-0.57

ilon

-0.56

POSITIVE LOGITS

 aback

1.21

by

1.09

BY

0.92

by

0.91

 awake

0.79

ienced

0.75

By

0.74

By

0.73

 reeling

0.72

 ashore

0.72

Activations Density 0.215%

verbs that describe negative political outcomes

No Comments

No Known Activations