INDEX

Explanations

numerical values and related variable assignments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_22/width_16k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 ngang

-0.59

↵

-0.53

thalene

-0.50

Hochspringen

-0.49

setVerticalGroup

-0.48

ContentAsync

-0.48

 المعيارى

-0.47

 فريبيس

-0.47

жели

-0.47

 moved

-0.47

POSITIVE LOGITS

UnusedPrivate

0.63

ercice

0.55

 kaarangay

0.55

 Reſ

0.54

 Савез

0.51

 ↓,

0.50

"""

0.50

 Inſ

0.48

 ſch

0.47

ſelves

0.47

Activations Density 0.066%

numerical values and related variable assignments

No Comments

No Known Activations