INDEX

Explanations

numerical values or symbols within the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_36/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.36.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

MLLoader

-0.56

CppMethod

-0.56

mybatisplus

-0.52

 IERC

-0.49

ьаж

-0.47

⏟

-0.47

 OMITBAD

-0.47

RTGC

-0.46

tvguidetime

-0.46

POSITIVE LOGITS

0.45

true

0.44

<bos>

0.43

ordum

0.42

0.40

________________

0.40

0.39

<i>

0.39

sort

0.38

Activations Density 1.203%

numerical values or symbols within the text

No Comments

No Known Activations