INDEX

Explanations

references to biological or biochemical terms and structures, particularly those indicating proteins or antibodies

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

starting with "IB", "OB", or "EB"

np_acts-logits-general · gemini-2.0-flash

IB, ob, EB, parab, xb

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

uple

-0.55

 integral

-0.52

DEFINE

-0.51

ActionEvent

-0.51

 Meksiku

-0.50

er

-0.49

 cata

-0.49

atguigu

-0.48

/******/

-0.47

POSITIVE LOGITS

SIB

0.64

Sib

0.59

bing

0.59

ecause

0.58

usiness

0.57

oneofs

0.56

bed

0.56

idopsis

0.54

twimg

0.54

neb

0.53

Activations Density 0.451%

references to biological or biochemical terms and structures, particularly those indicating proteins or antibodies

starting with "IB", "OB", or "EB"

IB, ob, EB, parab, xb

No Comments

No Known Activations