INDEX

Explanations

references to authorship and attribution in reporting

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

anel

-0.15

skirts

-0.15

 Damen

-0.14

nds

-0.14

elow

-0.14

gil

-0.14

iph

-0.14

zs

-0.14

ëŁ´

-0.14

POSITIVE LOGITS

 Reuters

0.22

 Editing

0.19

.Reporting

0.18

Reuters

0.18

 editing

0.16

 REUTERS

0.15

olon

0.15

Ã¶yle

0.15

Reporting

0.15

raft

0.15

Activations Density 0.004%

references to authorship and attribution in reporting

No Comments

No Known Activations