INDEX

Explanations

phrases indicating conditional or hypothetical situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

oload

-0.16

egrator

-0.14

ENTA

-0.13

isko

-0.13

issen

-0.13

mayan

-0.13

ndern

-0.13

utters

-0.13

ctp

-0.13

PFN

-0.13

POSITIVE LOGITS

 thing

1.38

thing

1.12

 Thing

1.07

Thing

0.97

 cosa

0.80

 things

0.79

THING

0.71

 coisa

0.71

(thing

0.71

things

0.66

Activations Density 0.175%

phrases indicating conditional or hypothetical situations

No Comments

No Known Activations