INDEX

Explanations

repeated mentions of specific characters or names in interactions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

æĽ

-0.16

ÑĨÐµÐ½

-0.16

bolt

-0.16

kud

-0.15

 scrim

-0.14

ertools

-0.14

.Utc

-0.14

/Graphics

-0.14

Ð¾ÑĢÑĭ

-0.14

fab

-0.14

POSITIVE LOGITS

ames

0.32

agh

0.31

aj

0.31

ajs

0.30

aja

0.29

atan

0.28

akes

0.28

ajas

0.28

avi

0.27

ishi

0.27

Activations Density 0.032%

repeated mentions of specific characters or names in interactions

No Comments

No Known Activations