INDEX

Explanations

negations and references to deception

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

icie

-0.15

Ð¾Ð¼ÐµÑĤ

-0.14

ãĥĩãĤ£ãĤ¢

-0.14

lero

-0.14

achable

-0.14

 dialog

-0.14

 tongues

-0.14

oÅĻ

-0.14

Uvs

-0.14

skirts

-0.14

POSITIVE LOGITS

 Thor

0.15

ipient

0.15

 harms

0.14

.BorderStyle

0.14

 ÑģÐ²

0.14

esson

0.14

aticon

0.13

çŃĴ

0.13

 tÃ©c

0.13

å»ł

0.13

Activations Density 0.001%

negations and references to deception

No Comments

No Known Activations

negations and references to deception

No Comments

No Known Activations