INDEX

Explanations

various approaches or methodologies in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

imity

-0.17

ucc

-0.15

GLOBALS

-0.15

udos

-0.15

ÐµÐ¶

-0.15

egal

-0.14

erral

-0.14

.nextToken

-0.14

Ð½ÐµÐ½

-0.14

aket

-0.14

POSITIVE LOGITS

able

0.15

 approaches

0.15

inz

0.14

 approach

0.14

Startup

0.14

anton

0.14

åľ°åĮº

0.14

_md

0.14

ise

0.13

Activations Density 0.016%

various approaches or methodologies in different contexts

No Comments

No Known Activations