INDEX

Explanations

coded references and numerical identifiers in structured data

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ylv

-0.15

nyder

-0.15

alc

-0.14

Ð¾ÑīÐ¸

-0.14

igg

-0.13

reen

-0.13

Positions

-0.13

?action

-0.13

 flat

-0.13

POSITIVE LOGITS

nier

0.18

zos

0.17

nost

0.15

 Gerry

0.15

utan

0.14

nem

0.14

ackle

0.14

ãĥ³ãĥĳ

0.13

anon

0.13

inand

0.13

Activations Density 0.331%

coded references and numerical identifiers in structured data

No Comments

No Known Activations