INDEX

Explanations

terms related to processes or measurements in scientific or technical contexts

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

rate constant calculation

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

msgTypes

-0.63

 CreateTagHelper

-0.58

 AssemblyCompany

-0.54

principalColumn

-0.54

 ProtoMessage

-0.50

MLLoader

-0.50

EndContext

-0.48

UnusedPrivate

-0.47

HasForeignKey

-0.46

rungsseite

-0.46

POSITIVE LOGITS

を行う

0.39

TabStop

0.38

Sr

0.37

 esterni

0.36

Trit

0.35

 transi

0.35

Filmographie

0.35

=")

0.34

按

0.33

按

0.33

Activations Density 0.039%

terms related to processes or measurements in scientific or technical contexts

rate constant calculation

No Comments

No Known Activations