INDEX

Explanations

references to computational operations or commands

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

operation

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

 confiable

-0.50

 Ambiental

-0.49

 abundancia

-0.48

 contemporánea

-0.46

sidemargin

-0.46

 bright

-0.46

 trustworthiness

-0.46

 templado

-0.45

 zasi

-0.45

 brightest

-0.45

POSITIVE LOGITS

 Operation

0.96

 operation

0.91

Operation

0.89

 Operations

0.85

 operations

0.84

Operations

0.83

operation

0.81

操作

0.77

operations

0.77

 oper

0.74

Activations Density 0.016%

references to computational operations or commands

operation

No Comments

No Known Activations