INDEX

Explanations

references to specific names or terms related to alignment and organization

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 27-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_27/width_16k/average_l0_118

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.27.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

endphp

-0.54

 kasarigan

-0.53

MockMvc

-0.52

pgterms

-0.51

adaptiveStyles

-0.51

RegressionTest

-0.51

expandindo

-0.50

 MainAxisSize

-0.50

styleType

-0.50

parsedMessage

-0.49

POSITIVE LOGITS

Alignment

0.62

 Alignment

0.61

 Alig

0.60

 align

0.57

 aligning

0.56

 alig

0.55

 ALIGN

0.54

 aligns

0.54

 Align

0.53

Align

0.52

Activations Density 0.158%

references to specific names or terms related to alignment and organization

No Comments

No Known Activations