INDEX

Explanations

terms related to data rows in a technical or programming context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_0/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.0.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

Rows

-1.01

tual

-0.68

PACKET

-0.64

 Rows

-0.62

Columns

-0.61

Ake

-0.60

valis

-0.60

 getContent

-0.59

 prime

-0.59

aderie

-0.59

POSITIVE LOGITS

 CreateTagHelper

0.82

Personendaten

0.77

ness

0.68

 跳转至

0.68

 INTERESAR

0.67

rasında

0.66

?”.

0.65

),”

0.63

ValueStyle

0.62

?”,

0.61

Activations Density 0.017%

terms related to data rows in a technical or programming context

No Comments

No Known Activations