INDEX

Explanations

references to visual elements such as figures and tables

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

ั้ง

-0.47

 niin

-0.46

 memb

-0.44

üks

-0.42

oustics

-0.41

PreAuthorize

-0.41

cuma

-0.41

 henne

-0.40

ctics

-0.39

isChecked

-0.39

POSITIVE LOGITS

 snippetHide

0.88

 surla

0.84

 дописавши

0.76

 kasarigan

0.75

FormTagHelper

0.74

rungsseite

0.72

 initComponents

0.69

+#+#

0.68

 архивлан

0.67

RegressionTest

0.66

Activations Density 0.521%

references to visual elements such as figures and tables

No Comments

No Known Activations