INDEX

Explanations

expressions related to thought processes and reflections

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

er

-0.73

drücken

-0.69

FilterChain

-0.69

-0.67

legungen

-0.67

-0.66

тельству

-0.65

ers

-0.64

anlagen

-0.64

рыва

-0.64

POSITIVE LOGITS

 Thought

1.31

 THOUGHT

1.30

thought

1.24

Thought

1.19

 thought

1.09

thoughts

1.06

 thoughts

1.02

SOT

0.98

 Thoughts

0.98

Manbalar

0.86

Activations Density 0.059%

expressions related to thought processes and reflections

No Comments

No Known Activations