INDEX

Explanations

formatted elements in programming or mathematical expressions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

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

RegressionTest

-0.97

aarrggbb

-0.87

 kasarigan

-0.85

########.

-0.79

 CreateTagHelper

-0.77

RenderAtEndOf

-0.72

 betweenstory

-0.69

 متعلقه

-0.66

 Wikimedijinoj

-0.65

 مرئيه

-0.64

POSITIVE LOGITS

NUMX

0.94

 ་་

0.84



0.83

\<^

0.78

ousand

0.77

ſelves

0.76

NESDAY

0.75

iffion

0.74

$_"

0.73

hematical

0.73

Activations Density 0.529%

formatted elements in programming or mathematical expressions

No Comments

No Known Activations