INDEX

Explanations

phrases that refer to various situations or contexts, often indicating a level of seriousness or complexity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

ends

-0.19

andra

-0.17

endale

-0.16

endas

-0.16

ieves

-0.16

enda

-0.15

esian

-0.15

ongyang

-0.15

Ð¾Ð²Ð¸

-0.15

aim

-0.15

POSITIVE LOGITS

ally

0.36

ality

0.25

als

0.23

nal

0.22

quo

0.22

 circumstances

0.21

 faced

0.20

/context

0.20

ALLY

0.20

nement

0.20

Activations Density 0.042%

phrases that refer to various situations or contexts, often indicating a level of seriousness or complexity

No Comments

No Known Activations