INDEX

Explanations

names and numerical identifiers, likely referencing people or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

ignon

-0.17

esty

-0.16

verse

-0.15

ÅĻi

-0.15

vala

-0.15

riers

-0.14

loff

-0.14

bst

-0.14

ilio

-0.14

 PdfP

-0.14

POSITIVE LOGITS

olland

0.14

 íĿ¬

0.14

Î¹ÎºÎ±

0.14

ilarity

0.13

fol

0.13

é©¾

0.13

QC

0.13

 Attempt

0.13

 amateur

0.13

ISC

0.13

Activations Density 0.011%

names and numerical identifiers, likely referencing people or entities

No Comments

No Known Activations