INDEX

Explanations

instances of place names and film titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

airo

-0.16

ÐµÑĢÐ¶

-0.15

Cha

-0.15

 Shan

-0.15

 ConnectionState

-0.15

å¨±ä¹Ĳ

-0.14

atra

-0.14

(fullfile

-0.14

çĦ

-0.14

 Rock

-0.13

POSITIVE LOGITS

ucker

0.16

atrix

0.15

ewhat

0.14

ameda

0.14

incer

0.14

ÏģÎ¹Î±

0.14

Ã¼re

0.14

bolt

0.14

isches

0.13

bows

0.13

Activations Density 0.063%

instances of place names and film titles

No Comments

No Known Activations