INDEX

Explanations

expressions related to actions, intentions, and moral judgments

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

doing and accomplishments

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 resourceCulture

-0.36

 CreateTagHelper

-0.35

 createSlice

-0.35

 gydy

-0.33

Revenir

-0.33

fjspx

-0.33

 žaid

-0.32

 llanto

-0.31

ėjimo

-0.31

 användas

-0.31

POSITIVE LOGITS

 done

0.93

done

0.83

 undone

0.77

 Done

0.73

Done

0.72

 accomplished

0.70

 dilakukan

0.70

 performed

0.67

 accomplish

0.66

 Performed

0.65

Activations Density 0.071%

expressions related to actions, intentions, and moral judgments

doing and accomplishments

No Comments

No Known Activations