INDEX

Explanations

commonality and high-frequency terms related to location and direction

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

addon

-0.15

ampa

-0.14

open

-0.14

 ÙħØ¹Ø¯

-0.14

endra

-0.14

mui

-0.14

ja

-0.14

inos

-0.14

ãĥ¼ãĥĩ

-0.14

 Institutions

-0.13

POSITIVE LOGITS

 local

0.20

ahy

0.16

local

0.15

 Jord

0.15

usercontent

0.15

.local

0.15

 AraÅŁtÄ±r

0.15

-local

0.14

acher

0.14

éĻĦè¿ĳ

0.14

Activations Density 0.006%

commonality and high-frequency terms related to location and direction

No Comments

No Known Activations