INDEX

Explanations

ellipses or pauses in text, indicating omission or trailing thoughts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

']))

-0.54

 Schenk

-0.54

Mint

-0.53

 sliver

-0.53

 onyx

-0.52

 Mette

-0.50

 nuance

-0.50

}}^

-0.50

']

-0.49

 Slip

-0.49

POSITIVE LOGITS

.....

0.57

ooooo

0.56

….”

0.56

----

0.54

 ....

0.54

....

0.53

 ----

0.52

!!!!

0.52

)....

0.52

...."

0.52

Activations Density 0.015%

ellipses or pauses in text, indicating omission or trailing thoughts

No Comments

No Known Activations