INDEX

Explanations

references to numerical values or quantities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

antan

-0.20

 outsider

-0.16

&a

-0.15

ied

-0.15

uyáº¿t

-0.15

iams

-0.15

zin

-0.14

-outs

-0.14

OUTPUT

-0.14

sgiving

-0.14

POSITIVE LOGITS

 nowhere

0.36

 bounds

0.30

 reach

0.27

 sight

0.25

 Bounds

0.22

reach

0.21

 pocket

0.21

bounds

0.20

 necessity

0.20

_bounds

0.20

Activations Density 0.050%

references to numerical values or quantities

No Comments

No Known Activations