INDEX

Explanations

references to time or dates, particularly the word "of" followed by numbers, indicating specific years or time frames

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

kok

-0.19

 Kraj

-0.15

 slip

-0.14

egt

-0.14

/us

-0.14

ubb

-0.14

cob

-0.13

ì§¸

-0.13

orch

-0.13

fold

-0.13

POSITIVE LOGITS

AttributeName

0.16

Latch

0.14

keh

0.14

riel

0.14

opia

0.14

 âĸ²

0.14

erset

0.14

_latest

0.14

omo

0.14

0.13

Activations Density 0.014%

references to time or dates, particularly the word "of" followed by numbers, indicating specific years or time frames

No Comments

No Known Activations