INDEX

Explanations

keywords and phrases related to actions and intentions, particularly involving deception or retrieval

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-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.hook_resid_post

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

NOPQRST

-0.67

CloseOperation

-0.61

 الحره

-0.61

 OFDb

-0.59

withIdentifier

-0.59

DoubleQuotes

-0.59

ؤلاء

-0.58

iança

-0.57

qrstuvwxyz

-0.57

uição

-0.53

POSITIVE LOGITS

it

4.10

它

2.43

It

2.42

 оно

2.35

It

2.35

nó

2.13

it

2.00

มัน

1.97

它

1.95

它是

1.73

Activations Density 4.273%

keywords and phrases related to actions and intentions, particularly involving deception or retrieval

No Comments

No Known Activations