INDEX

Explanations

words related to "duf" and its variations in context, indicating a focus on specific names or terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

er

-0.28

eru

-0.24

-0.21

erer

-0.21

Øª

-0.19

erot

-0.18

ig

-0.18

IG

-0.17

ero

-0.17

erse

-0.17

POSITIVE LOGITS

eteria

0.25

ords

0.24

uegos

0.22

leur

0.22

ORD

0.21

sky

0.20

amiliar

0.20

onso

0.20

eguard

0.20

raz

0.19

Activations Density 0.055%

words related to "duf" and its variations in context, indicating a focus on specific names or terms

No Comments

No Known Activations