INDEX

Explanations

references to time periods and significant milestones

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

isas

-0.18

habi

-0.16

æĥħ

-0.15

upe

-0.15

èĳī

-0.14

ÐµÐ·Ð¿ÐµÑĩ

-0.14

.CompareTo

-0.14

oster

-0.14

gren

-0.13

à¸Ńà¸ĩà¸Ħ

-0.13

POSITIVE LOGITS

 alive

0.22

 Alive

0.19

alive

0.17

of

0.16

oload

0.15

 vivo

0.14

ï»¿#

0.14

 elapsed

0.14

 into

0.14

ird

0.14

Activations Density 0.048%

references to time periods and significant milestones

No Comments

No Known Activations