INDEX

Explanations

interrogative and conditional phrases related to questioning or uncertainty

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

dera

-0.16

.Management

-0.15

mdb

-0.15

_HI

-0.14

_typeof

-0.14

 Ú¯Ø±ÙģØªÙĩ

-0.14

Ø§Ø¯Ùĩ

-0.14

arÃŃa

-0.13

dere

-0.13

cona

-0.13

POSITIVE LOGITS

did

1.13

Did

1.05

Did

1.01

did

0.99

DID

0.87

.did

0.80

 didn

0.71

 Didn

0.64

 didnt

0.63

didn

0.59

Activations Density 0.317%

interrogative and conditional phrases related to questioning or uncertainty

No Comments

No Known Activations