INDEX

Explanations

references to theorems and their proofs in mathematical discussions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

leton

-0.15

oller

-0.14

/Grid

-0.14

/material

-0.14

umas

-0.14

inston

-0.14

/body

-0.13

Gib

-0.13

bir

-0.13

POSITIVE LOGITS

arkin

0.14

_lineno

0.14

LOCKS

0.14

ervas

0.14

Ð¾ÐºÐ¾Ð»

0.14

.px

0.14

æ¨

0.13

æĪĲç«ĭ

0.13

 viol

0.13

mlx

0.13

Activations Density 0.103%

references to theorems and their proofs in mathematical discussions

No Comments

No Known Activations