INDEX

Explanations

technical terms and references related to academic or legal contexts

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-mlp/layer_22/width_16k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.hook_mlp_out

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 &___

-0.75

RenderAtEndOf

-0.60

 betweenstory

-0.58

 gyhoeddwyd

-0.56

ckner

-0.55

posedge

-0.54

GIVEREF

-0.54

 незавершена

-0.54

 EXISTS

-0.54

 AssemblyTitle

-0.54

POSITIVE LOGITS

gemeen

0.45

ValueStyle

0.45

 cord

0.44

ERT

0.44

hu

0.43

xlarge

0.42

()?;

0.42

ATN

0.41

iyi

0.41

rophi

0.41

Activations Density 2.097%

technical terms and references related to academic or legal contexts

No Comments

No Known Activations