INDEX

Explanations

phrases indicating ability, responsibility, and memory

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

Follows a modal verb

np_acts-logits-general · gemini-2.0-flash

modal verb + action

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.31.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

 usarlo

-0.63

 hacerlo

-0.63

httphttps

-0.59

دانشنامهٔ

-0.58

 незавершена

-0.52

 korzyst

-0.50

 Behör

-0.49

utilisons

-0.48

 bowiem

-0.48

 šķ

-0.46

POSITIVE LOGITS

_:

0.47

stdc

0.45

TextHelper

0.42

NegativeButton

0.41

 Италијани

0.41

tagHelperRunner

0.41

VolleyError

0.40

☸

0.40

蚪

0.40

}=\{

0.40

Activations Density 0.263%

phrases indicating ability, responsibility, and memory

Follows a modal verb

modal verb + action

No Comments

No Known Activations