INDEX

Explanations

sequences of function calls or array manipulations

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

itness

-0.16

urma

-0.16

wyn

-0.16

isin

-0.14

heap

-0.14

BIN

-0.14

estroy

-0.14

peak

-0.14

 flag

-0.14

POSITIVE LOGITS

icz

0.20

ses

0.16

>Error

0.16

*****/↵

0.15

MMdd

0.14

/settings

0.14

åŃ

0.14

dings

0.13

esor

0.13

 addCriterion

0.13

Activations Density 0.329%

sequences of function calls or array manipulations

No Comments

No Known Activations