INDEX

Explanations

references to past events or temporal markers in the context of observations or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ellido

-0.15

Ø±Ø§ÙĨ

-0.15

Intensity

-0.15

issing

-0.14

vio

-0.14

lington

-0.14

ushima

-0.14

 à¤Ĩà¤§

-0.13

igel

-0.13

 upgrade

-0.13

POSITIVE LOGITS

STALL

0.15

 CreateMap

0.15

ÑģÑĤÐ²

0.15

0.14

erv

0.14

ylon

0.14

arrass

0.14

 Gang

0.14

edia

0.13

Activations Density 0.174%

references to past events or temporal markers in the context of observations or conditions

No Comments

No Known Activations