INDEX

Explanations

references to events, timings, and preparations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

onth

-0.15

ULO

-0.14

iese

-0.14

áº£y

-0.13

Ð¿Ð¾Ñĩ

-0.13

 util

-0.13

anie

-0.13

 occasion

-0.13

ets

-0.13

iminal

-0.13

POSITIVE LOGITS

 orientation

0.25

 nationals

0.23

 orientations

0.23

 Orientation

0.21

 practice

0.20

 recess

0.19

orientation

0.19

Orientation

0.19

 voir

0.17

 Round

0.17

Activations Density 0.475%

references to events, timings, and preparations

No Comments

No Known Activations