INDEX

Explanations

names of individuals, particularly in legal or academic contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ibName

-0.14

 èĪ

-0.14

(?:

-0.14

æĪ¶

-0.14

_fds

-0.14

eters

-0.14

eldon

-0.13

amina

-0.13

miss

-0.13

erts

-0.13

POSITIVE LOGITS

ount

0.15

ijken

0.15

usb

0.14

usz

0.14

ounters

0.14

utas

0.14

bum

0.14

0.13

à¥°

0.13

olumn

0.13

Activations Density 0.126%

names of individuals, particularly in legal or academic contexts

No Comments

No Known Activations