INDEX

Explanations

the word "by" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 39-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_39/width_16k/average_l0_131

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.39.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

 helst

-0.44

ulihan

-0.39

Composable

-0.39

SuppressLint

-0.39

UnusedPrivate

-0.38

horabuena

-0.36

 utafitiHapana

-0.35

HomeAsUpEnabled

-0.35

VolleyError

-0.34

✭

-0.34

POSITIVE LOGITS

 means

1.26

 virtue

1.16

way

1.04

standers

0.99

 dint

0.98

stander

0.93

 MEANS

0.90

stolic

0.85

means

0.83

products

0.83

Activations Density 0.449%

the word "by" in various contexts

No Comments

No Known Activations