INDEX

Explanations

coding-related terminology, particularly indicators of connection or relationships in data structures

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

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

TestClass

-0.71

 toJson

-0.60

Архівовано

-0.53

 referrerpolicy

-0.52

WithMany

-0.47

 BorderSide

-0.45

occasione

-0.41

 bruges

-0.40

BagConstraints

-0.38

 كومونز

-0.38

POSITIVE LOGITS

Strongly

2.41

 Strongly

2.23

strongly

1.77

 strongly

1.70

 fuertemente

1.20

 fortement

1.09

 fortemente

1.03

mbg

0.97

Strong

0.94

 Strong

0.91

Activations Density 0.000%

coding-related terminology, particularly indicators of connection or relationships in data structures

No Comments

No Known Activations