INDEX

Explanations

expressions related to logical reasoning and logical structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 légitime

-0.59

nyttet

-0.56

 feltro

-0.54

{{$

-0.54

didSet

-0.54

terior

-0.54

 Cortez

-0.53

 murale

-0.53

rasti

-0.52

">{{$

-0.52

POSITIVE LOGITS

 clear

1.37

clear

1.17

 Clear

1.04

Clear

1.00

 clearing

0.90

CLEAR

0.89

 clears

0.86

clearing

0.85

 cleared

0.85

AddTagHelper

0.84

Activations Density 0.538%

expressions related to logical reasoning and logical structures

No Comments

No Known Activations