INDEX

Explanations

personal pronouns followed by a verb indicating action or belief

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_fs768-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

768

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

 totality

-0.63

 electromagnetic

-0.61

PTS

-0.61

rules

-0.59

 Allied

-0.58

 Impact

-0.58

 whichever

-0.57

 relevance

-0.57

 Plaint

-0.57

 selves

-0.56

POSITIVE LOGITS

'm

1.67

've

1.41

 suppose

1.21

 guess

1.19

'll

1.19

'd

1.16

am

1.16

ggy

1.11

 presume

1.08

nex

1.07

Activations Density 1.862%

personal pronouns followed by a verb indicating action or belief

No Comments

No Known Activations