INDEX

Explanations

phrases related to inclusion or belonging

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

won

-0.15

rick

-0.15

:↵

-0.14

ovsky

-0.14

inning

-0.14

mes

-0.14

 Barang

-0.14

lest

-0.13

mes

-0.13

lette

-0.13

POSITIVE LOGITS

ãĥ¼ãĥĬ

0.15

ODB

0.15

ContentSize

0.14

hue

0.14

isci

0.14

iful

0.14

 GOODMAN

0.14

YST

0.13

fÃ¼h

0.13

VRT

0.13

Activations Density 0.099%

phrases related to inclusion or belonging

No Comments

No Known Activations