INDEX

Explanations

articles, particularly "a" and "an"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ulumi

-0.15

nish

-0.14

enin

-0.14

elin

-0.14

 alphabetical

-0.14

aket

-0.14

adier

-0.14

 smash

-0.14

 certain

-0.13

ogn

-0.13

POSITIVE LOGITS

Redux

0.17

0.16

etÃŃ

0.16

_fds

0.16

unuz

0.15

Ð°Ð½ÑĤ

0.15

.AddTransient

0.14

imization

0.14

uzz

0.14

 zÃ¡b

0.14

Activations Density 0.039%

articles, particularly "a" and "an"

No Comments

No Known Activations