INDEX

Explanations

numerical references or citations within the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

eric

-0.17

ives

-0.15

 Bowie

-0.14

Ð°ÑĤÑĭ

-0.14

uids

-0.14

å¯º

-0.14

å®ļ

-0.14

echan

-0.14

DS

-0.13

 stitch

-0.13

POSITIVE LOGITS

íĨłíĨł

0.18

à¸Ĺà¸²à¸Ļ

0.17

 MouseEventArgs

0.16

ARING

0.16

aring

0.15

ose

0.14

 Busty

0.14

olor

0.14

ÑĤÐ°Ð»

0.14

ãĥ¥

0.14

Activations Density 0.059%

numerical references or citations within the text

No Comments

No Known Activations