INDEX

Explanations

keywords related to things being eaten or consumed

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-mlp/layer_6/width_131k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

收到

-0.45

 invitation

-0.40

 sent

-0.35

 receive

-0.33

 rain

-0.33

 dispatched

-0.32

GenerationType

-0.32

 RECEIVE

-0.32

 received

-0.32

 RECEIVED

-0.32

POSITIVE LOGITS

 nakalista

0.74

GOTREF

0.64

SharedCtor

0.59

 Réponses

0.55

fjspx

0.55

FieldNumber

0.54

 betweenstory

0.54

Tembelea

0.53

 nahilalakip

0.51

出版年

0.50

Activations Density 0.030%

keywords related to things being eaten or consumed

No Comments

No Known Activations