INDEX

Explanations

references to the word "from" and phrases indicating origin or starting points

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

Following the word "from"

np_acts-logits-general · gemini-2.0-flash

from followed by origin or source

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 frattempo

-0.52

 elsewhere

-0.49

 Meanwhile

-0.47

 redor

-0.46

 auprès

-0.45

Meanwhile

-0.44

 أخرى

-0.43

 altrett

-0.42

LastError

-0.42

mayın

-0.42

POSITIVE LOGITS

sizeCache

0.90

#+#

0.88

AddTagHelper

0.88

 propOrder

0.87

 start

0.85

 cradle

0.85

 smallest

0.84

 conception

0.81

 inception

0.81

 beginning

0.80

Activations Density 0.134%

references to the word "from" and phrases indicating origin or starting points

Following the word "from"

from followed by origin or source

No Comments

No Known Activations