INDEX

Explanations

references to scientific evidence and real-world benefits

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

 indeed

-0.15

urved

-0.15

vais

-0.15

_simps

-0.15

inq

-0.14

andel

-0.14

rema

-0.14

izyon

-0.14

agne

-0.13

abetic

-0.13

POSITIVE LOGITS

 actual

0.36

 Actual

0.33

actual

0.33

Actual

0.29

å®ŀéĻħ

0.29

 real

0.29

real

0.25

 practical

0.25

(actual

0.24

å¯¦

0.24

Activations Density 0.010%

references to scientific evidence and real-world benefits

No Comments

No Known Activations