INDEX

Explanations

the word "with" and words that commonly appear near it like "great", "and"

oai_token-act-pair · gemini-2.0-flash

with

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Sucesor

-0.81

BufferException

-0.70

发表于

-0.63

 فريبيس

-0.58

LikeLike

-0.57

 وتسجيلات

-0.54

 EClass

-0.53

SupportActionBar

-0.52

__':

-0.52

ransition

-0.51

POSITIVE LOGITS

 great

0.61

great

0.56

 greatest

0.54

Argb

0.53

новниш

0.52

 tightness

0.51



0.50

GREAT

0.49

Климат

0.49

 Tears

0.48

Activations Density 1.335%

the word "with" and words that commonly appear near it like "great", "and"

with

No Comments

No Known Activations