INDEX

Explanations

variations of the root word "er" and its associated suffix forms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

orney

-0.15

Compat

-0.14

 personals

-0.14

azer

-0.14

ujte

-0.13

nels

-0.13

Ĺi

-0.13

imedia

-0.13

nam

-0.13

POSITIVE LOGITS

edor

0.16

etting

0.15

ACY

0.14

usu

0.14

nger

0.14

Äįin

0.14

ching

0.13

à¹ĥà¸Ļà¸Ĺ

0.13

ãĥĥãĥĪ

0.13

Activations Density 0.404%

variations of the root word "er" and its associated suffix forms

No Comments

No Known Activations