INDEX

Explanations

phrases that emphasize key conclusions or summaries

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

bottom line

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

findpost

-0.36

canestro

-0.34

ANDOM

-0.34

 Pender

-0.33

 sandwiches

-0.33

{@

-0.33

 flowers

-0.32

 candies

-0.32

 seeds

-0.32

稍

-0.32

POSITIVE LOGITS

Bottom

0.94

 Bottom

0.88

 bottom

0.82

bottom

0.81

 BOTTOM

0.79

BOTTOM

0.68

bottomRight

0.68

Line

0.64

Baseline

0.61

 Baseline

0.59

Activations Density 0.004%

phrases that emphasize key conclusions or summaries

bottom line

No Comments

No Known Activations