INDEX

Explanations

references to specific nationalities, cultures, and prominent institutions or ideologies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

orem

-0.33

/Sub

-0.21

/Set

-0.19

/Area

-0.18

/Branch

-0.17

suite

-0.16

-Day

-0.16

notated

-0.15

/Admin

-0.15

-Agent

-0.15

POSITIVE LOGITS

bidden

0.25

tempts

0.18

ventory

0.18

ufact

0.17

/or

0.17

nger

0.17

ses

0.17

:///

0.17

dependence

0.16

/english

0.16

Activations Density 0.580%

references to specific nationalities, cultures, and prominent institutions or ideologies

No Comments

No Known Activations