INDEX

Explanations

structured data or code representations, particularly related to graphics or display settings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

 חיצוניים

-0.54

 retrait

-0.51

ējās

-0.51

 odkazy

-0.51

protoc

-0.49

CGContext

-0.47

<tbody>

-0.47

 tensão

-0.46

यं

-0.44

 réf

-0.44

POSITIVE LOGITS

 Majefty

0.63

AnchorStyles

0.62

 ſta

0.61

 houſe

0.61

principalColumn

0.61

 Chriſt

0.60

 ſtate

0.60

 cauſe

0.60

 Houſe

0.60

 purpoſe

0.59

Activations Density 0.508%

structured data or code representations, particularly related to graphics or display settings

No Comments

No Known Activations