INDEX

Explanations

personal pronouns and possessive pronouns, specifically focusing on instances related to the speaker or writer

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

ories

-0.74

iston

-0.63

°

-0.61

 Centers

-0.60

 incorporation

-0.59

orie

-0.59

majority

-0.58

ftime

-0.57

profits

-0.57

nce

-0.56

POSITIVE LOGITS

 personally

1.06

lees

0.88

adows

0.87

atic

0.84

 somew

0.79

andering

0.78

 verbally

0.75

erk

0.74

 orally

0.73

 uncond

0.72

Activations Density 0.746%

personal pronouns and possessive pronouns, specifically focusing on instances related to the speaker or writer

No Comments

No Known Activations