INDEX

Explanations

the word "respective" and related terms indicating individual or distinct relationships

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ll

-0.15

ska

-0.15

book

-0.15

kah

-0.15

mong

-0.14

amentals

-0.14

ogo

-0.14

 superClass

-0.14

ibrary

-0.13

POSITIVE LOGITS

Ã¼zel

0.17

yz

0.16

erras

0.15

MENT

0.15

INGER

0.15

centroid

0.14

och

0.14

undles

0.14

cx

0.13

ivil

0.13

Activations Density 0.012%

the word "respective" and related terms indicating individual or distinct relationships

No Comments

No Known Activations