INDEX

Explanations

references to personal experience and familiarity

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

">//

-0.90

UnusedPrivate

-0.85

 Exactos

-0.81

 estekak

-0.70

]")]

-0.68

 مشين

-0.67

########.

-0.66

 tartalomajánló

-0.66

RenderAtEndOf

-0.65

!*\

-0.63

POSITIVE LOGITS

 BitSet

0.52

baomidou

0.50

 sexuales

0.49

bcrypt

0.46

ladesh

0.45

 orgasm

0.45

ctivo

0.43

 umum

0.43

rones

0.42

kasse

0.42

Activations Density 0.247%

references to personal experience and familiarity

No Comments

No Known Activations