INDEX

Explanations

phrases referring to potential outcomes or scenarios

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

possibility that

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

irchen

-0.49

 katholischen

-0.47

 Económica

-0.45

Measured

-0.43

 TÉCN

-0.42

 gesuchten

-0.42

 Geset

-0.42

mobileqq

-0.41

 usata

-0.41

Logged

-0.41

POSITIVE LOGITS

 Possibility

1.13

 possibility

1.10

possibility

1.05

Possibility

0.99

 Possibilities

0.94

 possibilities

0.87

 possib

0.76

 posibilidad

0.74

 likelihood

0.71

 Likelihood

0.70

Activations Density 0.013%

phrases referring to potential outcomes or scenarios

possibility that

No Comments

No Known Activations