INDEX

Explanations

the word "accordingly" and its variations to indicate a following logical conclusion or consequence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

oro

-0.15

usch

-0.15

emory

-0.15

ullo

-0.15

 Maze

-0.14

erot

-0.14

HOOK

-0.14

 Pear

-0.14

ophysical

-0.14

ovsky

-0.14

POSITIVE LOGITS

 traff

0.15

ounder

0.14

 abstract

0.14

 ØµØŃ

0.14

.Spring

0.14

 abrupt

0.14

_TAC

0.14

Ø±Ø¯

0.14

TAB

0.14

acente

0.14

Activations Density 0.002%

the word "accordingly" and its variations to indicate a following logical conclusion or consequence

No Comments

No Known Activations