INDEX

Explanations

programming language syntax involving parsing and responding to data

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

identification and conversion

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

ItemBackground

-0.68

SBATCH

-0.65

 كومونز

-0.63

 AssemblyCulture

-0.61

 SwitchCompat

-0.60

省市镇

-0.60

expandindo

-0.59

 '\\;'

-0.59

 <=",

-0.56

 queſta

-0.56

POSITIVE LOGITS

RRect

0.40

 convierten

0.37

 interpretación

0.37

 creado

0.35

 conversión

0.34

 AutoMapper

0.34

wartet

0.34

Вид

0.34

entuh

0.33

 transforming

0.33

Activations Density 0.007%

programming language syntax involving parsing and responding to data

identification and conversion

No Comments

No Known Activations