INDEX

Explanations

concepts related to uniqueness and individual features

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

ifu

-0.16

ÙĪÙĨØ§

-0.15

ORMAT

-0.14

eln

-0.14

imson

-0.14

atter

-0.14

ÑĮÐµ

-0.14

.Helper

-0.14

esser

-0.14

ika

-0.14

POSITIVE LOGITS

 respectively

0.27

 respective

0.23

 distinct

0.17

distinct

0.16

 ÑģÐ¾Ð¾ÑĤÐ²ÐµÑĤ

0.16

 Ulus

0.16

 nhau

0.15

ovÃŃ

0.15

åĪĨåĪ«

0.15

 Dance

0.14

Activations Density 0.108%

concepts related to uniqueness and individual features

No Comments

No Known Activations