INDEX

Explanations

references to structured data or numerical information presented within tables or lists

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

empo

-0.16

ilib

-0.15

eos

-0.14

ICENSE

-0.14

anches

-0.14

conto

-0.14

peg

-0.14

elect

-0.14

à¸£

-0.14

POSITIVE LOGITS

suite

0.15

nbsp

0.15

\/\/

0.14

itzer

0.14

raquo

0.14

Ð¾Ð·

0.14

ients

0.13

wel

0.13

agina

0.13

ispens

0.13

Activations Density 0.168%

references to structured data or numerical information presented within tables or lists

No Comments

No Known Activations