INDEX

Explanations

phrases related to political identities or affiliations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

veau

-0.15

ÙģØ§ÙĤ

-0.15

 Recap

-0.15

ordion

-0.14

ymoon

-0.14

ActionCreators

-0.14

Injection

-0.14

AXB

-0.14

Vig

-0.14

Č↵

-0.13

POSITIVE LOGITS

bas

0.16

ersh

0.15

ledge

0.14

onu

0.14

ĥĿ

0.14

ingle

0.14

 Brewer

0.14

ICA

0.14

erton

0.13

 sche

0.13

Activations Density 0.520%

phrases related to political identities or affiliations

No Comments

No Known Activations