INDEX

Explanations

phrases that indicate explanations or elucidations

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

 explain

-1.30

 explanation

-1.16

 explains

-1.14

 explaining

-1.13

 explained

-1.13

 explanations

-1.02

 Explain

-1.00

explain

-0.99

Explain

-0.99

 explicar

-0.98

POSITIVE LOGITS

TypedDataSet

0.82

 Roskov

0.80

^(@)

0.80

 الحره

0.77

UserScript

0.76

VIAF

0.73

 noDo

0.73

 تضيفلها

0.71

########.

0.69

unknownFields

0.69

Activations Density 0.256%

phrases that indicate explanations or elucidations

No Comments

No Known Activations