INDEX

Explanations

specific references to numerical data and external links or commands

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

urat

-0.16

amax

-0.15

ixed

-0.15

adas

-0.15

emat

-0.15

uf

-0.14

pis

-0.14

.nlm

-0.14

 Norton

-0.14

ieves

-0.14

POSITIVE LOGITS

éĬĢ

0.15

abra

0.15

 Maze

0.14

_Init

0.14

ãĤ¶ãĥ¼

0.13

Î¬Î²

0.13

apiro

0.13

 Orient

0.13

Enlarge

0.13

apo

0.13

Activations Density 0.003%

specific references to numerical data and external links or commands

No Comments

No Known Activations