INDEX

Explanations

closing brackets or punctuation marks in code or data structures

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

code closing brackets

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

<bos>

-0.58

 Velasco

-0.56

<sup>

-0.54

Atk

-0.54

sns

-0.51

Nass

-0.50

↵↵

-0.49

観て

-0.49

 perſ

-0.49

AFS

-0.48

POSITIVE LOGITS

'];

1.44

'];

1.26

"];

1.22

"];

1.09

}];

1.00

"];

0.96

'};

0.95

]]);

0.94

 "]";

0.94

)];

0.93

Activations Density 0.005%

closing brackets or punctuation marks in code or data structures

code closing brackets

No Comments

No Known Activations