INDEX

Explanations

programming-related functions and methods

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

function calls and expressions

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

delwed

-0.60

 CreateTagHelper

-0.52

 Chwiliwch

-0.51

 uovo

-0.50

rungsseite

-0.49

FormTagHelper

-0.48

principalColumn

-0.44

 numberOfRows

-0.43

 doInBackground

-0.43

TagMode

-0.43

POSITIVE LOGITS

nodoc

0.43

✨:

0.42

jsxFileName

0.41

先代

0.38

 estekak

0.37

orenje

0.37

であろう

0.36

ConfigureAwait

0.35

Портал

0.35

offsetof

0.35

Activations Density 0.048%

programming-related functions and methods

function calls and expressions

No Comments

No Known Activations