INDEX

Explanations

references to actions and observations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

OwnProperty

-0.16

ounder

-0.16

zes

-0.15

 observations

-0.15

 Bristol

-0.14

esian

-0.14

vert

-0.14

avra

-0.14

jiÅ¡Å¥

-0.14

 Observ

-0.14

POSITIVE LOGITS

 evidence

0.18

 faces

0.18

 signs

0.18

burgh

0.17

 face

0.16

 Signs

0.16

iface

0.15

Äįku

0.15

/sm

0.15

 results

0.15

Activations Density 0.093%

references to actions and observations

No Comments

No Known Activations