INDEX

Explanations

first-person and second-person pronouns indicating agency or ownership

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

icle

-0.71

 replication

-0.65

 casc

-0.64

 Citation

-0.63

 Overt

-0.62

 obscurity

-0.62

glers

-0.61

ibrary

-0.59

icles

-0.58

 entropy

-0.58

POSITIVE LOGITS

 intend

1.24

 intention

1.20

 intends

1.11

 intent

0.99

 intentions

0.98

 regretted

0.93

 intending

0.88

 intended

0.84

 liking

0.83

 plans

0.81

Activations Density 0.221%

first-person and second-person pronouns indicating agency or ownership

No Comments

No Known Activations