INDEX

Explanations

tokens that indicate structured formatting or elements in programming contexts

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

Junk characters and non-English text

np_acts-logits-general · gemini-2.0-flash

specific numbers and punctuation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

transQ

-0.82

Hentet

-0.80

 صوتيه

-0.79

󠁢

-0.74

 "}";

-0.67

 "]";

-0.66

AddTagHelper

-0.62

}),

-0.61

Legături

-0.61

-0.59

POSITIVE LOGITS

OOTDTY

0.61

gebob

0.60

tvguidetime

0.59

complexContent

0.59

Földrajzportál

0.57

허

0.56

mobileqq

0.55

AsStream

0.55

 librement

0.54

qrstuvwxyz

0.54

Activations Density 0.112%

tokens that indicate structured formatting or elements in programming contexts

Junk characters and non-English text

specific numbers and punctuation

No Comments

No Known Activations