INDEX

Explanations

numerical references or data points

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

academic citations or references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.16.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

ChildScrollView

-1.02

 تكبرها

-0.78

Portály

-0.78

sidemargin

-0.73

 فريبيس

-0.73

 незавершена

-0.70

TestTools

-0.70

 cucchia

-0.69

Parcelize

-0.68

LookAnd

-0.66

POSITIVE LOGITS

motic

0.49

لمانيا

0.47

曜日

0.47

<eos>

0.46

:].

0.46

kedés

0.45

]['

0.44

/*!

0.44

</em>

0.44

0.43

Activations Density 0.011%

numerical references or data points

academic citations or references

No Comments

No Known Activations