INDEX

Explanations

indicators of emotional regulation and cognitive evaluation

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

 Fiesta

-0.79

 perks

-0.73

 Patriot

-0.72

ulhu

-0.72

 Benny

-0.72

 chops

-0.71

 stunts

-0.71

Wiz

-0.70

 Borders

-0.70

 Mama

-0.70

POSITIVE LOGITS

derived

1.28

biased

1.20

induced

1.17

mediated

1.16

treatment

1.16

directed

1.16

treated

1.15

negative

1.15

specific

1.15

adapt

1.14

Activations Density 0.067%

indicators of emotional regulation and cognitive evaluation

No Comments

No Known Activations