INDEX

Explanations

portions of text related to quantity or totality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

ordial

-0.17

Ð¾Ð¼ÐµÐ½

-0.15

ÎŃÏģÎ±

-0.15

ICC

-0.15

ryn

-0.15

mong

-0.14

 Expression

-0.14

ourke

-0.14

 rhyme

-0.14

icum

-0.14

POSITIVE LOGITS

geme

0.17

/all

0.17

igator

0.17

ingers

0.15

uded

0.15

LLL

0.15

usive

0.15

uv

0.15

iance

0.15

resco

0.14

Activations Density 0.099%

portions of text related to quantity or totality

No Comments

No Known Activations