INDEX

Explanations

numerical values and their associated context

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

 intent

-0.17

dra

-0.14

ierge

-0.14

 Aires

-0.14

 Tanner

-0.14

pedia

-0.13

vara

-0.13

ho

-0.13

 humanities

-0.13

 conscience

-0.13

POSITIVE LOGITS

neau

0.17

stoi

0.16

endregion

0.16

.lucene

0.15

ToWorld

0.15

åĤ

0.14

Ð»Ð¸Ñħ

0.14

.flink

0.14

.LoadScene

0.14

stream

0.14

Activations Density 0.191%

numerical values and their associated context

No Comments

No Known Activations