INDEX

Explanations

names of legal cases or references

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

prech

-0.17

ebin

-0.16

embed

-0.15

bourg

-0.15

udies

-0.15

mÃ¼r

-0.14

ÑĢÐ¾ÑĤ

-0.14

 embed

-0.14

ÙĨØ§ÙĨ

-0.14

imitive

-0.14

POSITIVE LOGITS

ay

0.14

_MATH

0.13

HL

0.13

arrow

0.13

 Rescue

0.13

ep

0.13

emoc

0.12

du

0.12

Activations Density 0.020%

names of legal cases or references

No Comments

No Known Activations