INDEX

Explanations

start tokens indicating the beginning of sentences or paragraphs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_11/width_16k/average_l0_118

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

}}%

-0.41

 Olympedia

-0.38

 ruine

-0.37

↵↵↵

-0.35

 PLWABN

-0.34

<eos>

-0.33

}}}}

-0.33

 Zorn

-0.33

bParam

-0.32

↵↵↵↵

-0.31

POSITIVE LOGITS



0.66

0.61

 itchy

0.59

 handcuffs

0.59

 ${({

0.58

iſchen

0.57

 fashiola

0.56



0.56

掟

0.56

 treachery

0.56

Activations Density 0.082%

start tokens indicating the beginning of sentences or paragraphs

No Comments

No Known Activations