INDEX

Explanations

text related to precise instructions or guidelines for a procedure

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

Objects being inserted/drilled into things

np_acts-logits-general · gemini-2.0-flash

needle insertion and drilling

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

 שוליים

-0.60

 Lawson

-0.52

 """",

-0.51

 فريبيس

-0.50

 flaps

-0.49

 redistribution

-0.48

 Bianchi

-0.48

 scarves

-0.48

 parcels

-0.46

 envelopes

-0.46

POSITIVE LOGITS

 drilling

0.98

 inserted

0.96

 needle

0.95

Drilling

0.95

 pierce

0.94

 drilled

0.93

 needles

0.93

 insertion

0.92

 drill

0.92

 Drill

0.91

Activations Density 0.527%

text related to precise instructions or guidelines for a procedure

Objects being inserted/drilled into things

needle insertion and drilling

No Comments

No Known Activations