INDEX

Explanations

numerical sequences or codes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 6-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_6/width_16k/average_l0_133

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.6.hook_mlp_out

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

脚注の使い方

-0.54

AddTagHelper

-0.52

 aikaa

-0.49

 Parish

-0.49

Sal

-0.49

 '\\;'

-0.47

fohl

-0.47

 intptr

-0.47

fromnode

-0.47

inoza

-0.47

POSITIVE LOGITS

 itſelf

0.77

 pleaſure

0.73

 myſelf

0.73

 ſche

0.72

Efq

0.71

 Monfieur

0.71

 ſen

0.64

 ſeveral

0.62

 themſelves

0.62

 Theſe

0.61

Activations Density 0.529%

numerical sequences or codes

No Comments

No Known Activations