INDEX

Explanations

forms of the verb "to be" and related phrases indicating existence or description

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

'the following' introductions

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

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ſelf

-0.59

 ESPECIAL

-0.58

 ſta

-0.57

 faſt

-0.57

 ſte

-0.57

 houſe

-0.57

 rodríguez

-0.55

 sánchez

-0.55

ientí

-0.54

awtextra

-0.54

POSITIVE LOGITS

0.52

 following

0.48

 below

0.48

ex

0.47

 summaries

0.45

an

0.44

 unfolding

0.42

 summary

0.42

 ensuing

0.41

 detailed

0.40

Activations Density 0.021%

forms of the verb "to be" and related phrases indicating existence or description

'the following' introductions

No Comments

No Known Activations