INDEX

Explanations

phrases related to events occurring or actions taking place

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

LOPT

-0.21

enco

-0.19

vvm

-0.15

_ASSUME

-0.15

iyim

-0.14

$LANG

-0.14

jom

-0.14

unes

-0.14

ç£

-0.14

ìĤ¬ëŀĳ

-0.14

POSITIVE LOGITS

 place

0.69

place

0.51

.place

0.42

 Place

0.42

 places

0.42

_place

0.40

Place

0.40

 PLACE

0.40

-place

0.40

 lugar

0.38

Activations Density 0.014%

phrases related to events occurring or actions taking place

No Comments

No Known Activations