INDEX

Explanations

references to loading and associated actions or operations

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

the word "load"

np_acts-logits-general · gemini-2.0-flash

load and its contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 Enix

-0.64

+#+#

-0.60

 Svensk

-0.59

 示例

-0.58

Abitanti

-0.58

 bezel

-0.57

Vex

-0.57

WriteLiteral

-0.56

 getArguments

-0.56

esModule

-0.56

POSITIVE LOGITS

 Load

1.17

 load

1.15

LOAD

1.13

 LOAD

1.12

loads

1.09

 loads

1.09

 Loads

1.08

load

1.08

Load

1.06

unload

1.00

Activations Density 0.115%

references to loading and associated actions or operations

the word "load"

load and its contexts

No Comments

No Known Activations