INDEX

Explanations

occurrences of the word "that" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_36/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.36.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

yntaxException

-0.64

 ब्रेकडाउन

-0.49

 habet

-0.49

 represents

-0.47

 wydaje

-0.46

 znajduje

-0.44

 nemá

-0.44

 może

-0.43

 consists

-0.43

 symbolizes

-0.43

POSITIVE LOGITS

is

1.50

Is

1.38

Is

1.23

is

1.23

IS

1.02

 getIs

0.79

isl

0.74

IS

0.73

 است

0.69

isk

0.69

Activations Density 0.549%

occurrences of the word "that" in various contexts

No Comments

No Known Activations