INDEX

Explanations

instances of specific names, references, or elements relevant to various contexts, particularly in academia, research, and cultural references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 2-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 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

<bos>

-0.42

-0.41

encodeWith

-0.39

<eos>

-0.39

-0.35

-0.34

ch

-0.34

 inspection

-0.34

 peak

-0.34

-0.33

POSITIVE LOGITS

Autoritní

0.73

rungsseite

0.72

principalTable

0.71

########.

0.70

 незавершена

0.67

tvguidetime

0.66

 autorytatywna

0.65

afficheront

0.65

̈́

0.65

 ब्रेकडाउन

0.64

Activations Density 1.994%

instances of specific names, references, or elements relevant to various contexts, particularly in academia, research, and cultural references

No Comments

No Known Activations