INDEX

Explanations

frequent and recurring patterns or structures in texts

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

code keywords and syntax

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

 betweenstory

-1.00

]")]

-0.88

ThroughAttribute

-0.83

 XCTest

-0.78

iſchen

-0.77

jsonwebtoken

-0.77

RTEE

-0.77

ſſung

-0.77

msgTypes

-0.76

niſſe

-0.76

POSITIVE LOGITS

1.15

0.96

0.81

0.78

0.74

0.73

0.70

Activations Density 0.033%

frequent and recurring patterns or structures in texts

code keywords and syntax

No Comments

No Known Activations