INDEX

Explanations

references to the name "Michael" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

age

-0.17

-long

-0.14

 misunder

-0.14

ong

-0.14

allon

-0.14

long

-0.14

>Delete

-0.14

æ¶²

-0.13

byt

-0.13

 gÃ¶re

-0.13

POSITIVE LOGITS

thus

0.15

onse

0.14

kami

0.14

kola

0.14

peace

0.14

pell

0.14

apas

0.14

hdl

0.13

atonin

0.13

oje

0.13

Activations Density 0.014%

references to the name "Michael" in various contexts

No Comments

No Known Activations