INDEX

Explanations

references to titles and names related to works

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

aarrggbb

-0.98

complexContent

-0.91

 שוליים

-0.81

PMailer

-0.76

 صوتيه

-0.74

RenderAtEndOf

-0.72

 initComponents

-0.70

SharedDtor

-0.70

İstinadlar

-0.69

Enllaços

-0.67

POSITIVE LOGITS

hi

0.49

السلام

0.45

 title

0.43

表

0.43

 stated

0.43

の通り

0.42

 above

0.42

 header

0.42

 colony

0.41

 dono

0.41

Activations Density 0.200%

references to titles and names related to works

No Comments

No Known Activations