INDEX

Explanations

instances of dialogue or discussion in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Zink

-0.72

 Cymru

-0.63

whelmed

-0.61

 Borden

-0.60

>>>>>>>>

-0.60

 plomb

-0.59

mal

-0.57

 Shelton

-0.55

ing

-0.54

entai

-0.53

POSITIVE LOGITS

 discuss

1.57

 Discuss

1.56

 discus

1.54

 discussion

1.53

Discuss

1.41

 discussing

1.41

 discussions

1.39

 discusses

1.39

discussion

1.39

discuss

1.35

Activations Density 0.068%

instances of dialogue or discussion in the text

No Comments

No Known Activations