INDEX

Explanations

instances of entities or attributes related to roles and classifications

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

edl

-0.16

cord

-0.15

prompt

-0.15

cid

-0.15

iven

-0.15

Latch

-0.14

Č↵

-0.14

iry

-0.13

otron

-0.13

cu

-0.13

POSITIVE LOGITS

?,

0.15

ï¼īãģ¯

0.14

by

0.14

ayet

0.14

McN

0.13

 fluid

0.13

ileceÄŁi

0.13

!,

0.13

iga

0.13

Activations Density 0.180%

instances of entities or attributes related to roles and classifications

No Comments

No Known Activations