INDEX

Explanations

instances of the word "once" suggesting repetition or a return to a previous state

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

another

-0.17

 Another

-0.17

arken

-0.16

aming

-0.16

wdx

-0.15

 another

-0.15

 Ð¿ÑĢÐ°Ð²Ð´Ð°

-0.15

antes

-0.15

Another

-0.15

bye

-0.15

POSITIVE LOGITS

 gain

0.32

gain

0.27

 Gain

0.25

ag

0.25

Gain

0.24

yet

0.23

AN

0.21

 gains

0.21

_gain

0.21

ag

0.21

Activations Density 0.018%

instances of the word "once" suggesting repetition or a return to a previous state

No Comments

No Known Activations