INDEX

Explanations

references to trees and tree-related terminology

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

trees

np_acts-logits-general · gemini-2.0-flash

tree and related compounds

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

MLLoader

-0.80

hdashline

-0.75

GGLE

-0.71

]-->

-0.71

 myſelf

-0.69

 createSlice

-0.68

 Jefus

-0.68

rators

-0.67

ñores

-0.67

✨:

-0.67

POSITIVE LOGITS

 branches

0.73

 Trees

0.71

 trunk

0.71

 trunks

0.69

 trees

0.69

tops

0.69

 branch

0.66

house

0.64

 stump

0.64

 Trunk

0.62

Activations Density 0.075%

references to trees and tree-related terminology

trees

tree and related compounds

No Comments

No Known Activations