INDEX

Explanations

references to research and findings within a scientific context

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

results and outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.16.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

 Henley

-0.50

DispatchToProps

-0.50

[]:

-0.49

 Gosling

-0.49

Define

-0.49

IntoConstraints

-0.49

 répar

-0.48

WithIdentifier

-0.48

 Guardiola

-0.47

roga

-0.46

POSITIVE LOGITS

 hasilnya

0.75

 ultimately

0.72

 melakukannya

0.71

 dabei

0.70

Ultimately

0.69

parsedMessage

0.69

 farlo

0.69

結果

0.67

 such

0.67

MessageTagHelper

0.66

Activations Density 0.220%

references to research and findings within a scientific context

results and outcomes

No Comments

No Known Activations