INDEX

Explanations

numerical data, particularly dates and any related counts or figures

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

BeginContext

-0.82

paksa

-0.82

 CURLOPT

-0.80

CURLOPT

-0.77

ardless

-0.72

mAh

-0.70

batis

-0.70

extAlignment

-0.69

 eslint

-0.68

lorette

-0.68

POSITIVE LOGITS

kem

0.57

 unknownFields

0.55

VENUE

0.52

ACTERS

0.51

 Knockout

0.50

Mako

0.48

̷

0.48

LookAnd

0.48

Ārējās

0.47



0.47

Activations Density 0.245%

numerical data, particularly dates and any related counts or figures

No Comments

No Known Activations