INDEX

Explanations

references to frames and their specifications in a technical context

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

frame

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 UserRepository

-0.60

hoeddwyd

-0.57

DotNetBar

-0.56

Personensuche

-0.56

oshiba

-0.50

tonsoft

-0.49

 testGet

-0.47

 Preferencias

-0.46

plish

-0.45

)++;

-0.44

POSITIVE LOGITS

 frame

2.22

frame

2.11

 Frame

2.09

Frame

1.98

 FRAME

1.90

 frames

1.88

FRAME

1.75

 Frames

1.72

frames

1.61

Frames

1.55

Activations Density 0.011%

references to frames and their specifications in a technical context

frame

No Comments

No Known Activations