INDEX

Explanations

words related to causation and argumentation

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

 Ø§ÙĦØªØ¬

-0.15

wl

-0.14

due

-0.14

qr

-0.14

WL

-0.14

ft

-0.14

Fut

-0.14

[".

-0.13

icontrol

-0.13

uess

-0.13

POSITIVE LOGITS

ivos

0.17

Ã¢ng

0.16

 Ð´Ð°Ð½Ð½Ñĭ

0.16

ãģĿ

0.14

.Clone

0.14

ä»£

0.14

ople

0.13

amura

0.13

reich

0.13

 buckle

0.13

Activations Density 0.052%

words related to causation and argumentation

No Comments

No Known Activations