INDEX

Explanations

terms related to boundaries and classifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 17-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_17/width_65k/average_l0_125

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.17.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 nakalista

-0.78

DebuggerNonUser

-0.76

 HasFactory

-0.70

 TextAlign

-0.65

談社

-0.62

 ©️

-0.61

stdafx

-0.60

setArguments

-0.59

 endwhile

-0.59

 FetchType

-0.59

POSITIVE LOGITS

 Betracht

0.51

calaure

0.49

icec

0.46

tup

0.42

ichel

0.42

zuf

0.41

hynch

0.41

payers

0.41

">'.$

0.40

Pab

0.40

Activations Density 0.308%

terms related to boundaries and classifications

No Comments

No Known Activations