INDEX

Explanations

phrases directing attention towards the reader or listener

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 35-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_35/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.35.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

JSONException

-0.47

 TextAppearance

-0.45

liceerd

-0.43

hols

-0.42

 HttpHeaders

-0.42

 JSONException

-0.42

これからも

-0.40

AnimationsModule

-0.39

NTB

-0.39

continue

-0.39

POSITIVE LOGITS

 already

0.66

 know

0.62

 ohnehin

0.61

Vidite

0.57

already

0.54

own

0.52

 normally

0.52

 KNOW

0.52

 owns

0.49

fromnode

0.49

Activations Density 0.738%

phrases directing attention towards the reader or listener

No Comments

No Known Activations