INDEX

Explanations

mentions of personal experiences

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

 electromagnetic

-0.63

 totality

-0.62

rules

-0.61

 relevance

-0.57

PTS

-0.56

 selves

-0.56

 uniform

-0.56

 Impact

-0.56

 Gale

-0.56

 tides

-0.55

POSITIVE LOGITS

'm

1.69

've

1.41

'll

1.19

 suppose

1.19

am

1.17

'd

1.17

 guess

1.14

ggy

1.09

ronic

1.07

 presume

1.06

Activations Density 0.823%

mentions of personal experiences

No Comments

No Known Activations