INDEX

Explanations

phrases emphasizing facts or beliefs

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 streng

-0.66

TPPStreamerBot

-0.64

ESE

-0.64

sung

-0.63

itsch

-0.62

wana

-0.61

nan

-0.61

 piping

-0.59

 bargain

-0.59

 lungs

-0.58

POSITIVE LOGITS

ually

1.22

uality

1.18

ional

1.15

orial

1.07

oids

0.90

itious

0.89

ual

0.87

uated

0.86

finding

0.83

oid

0.83

Activations Density 0.528%

phrases emphasizing facts or beliefs

No Comments

No Known Activations