INDEX

Explanations

code-related structures, particularly arrays, functions, and object manipulation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

bidden

-0.26

tempts

-0.20

ses

-0.17

nger

-0.17

rastructure

-0.17

DataGridViewTextBoxColumn

-0.17

efault

-0.15

/oct

-0.15

ArrayOf

-0.15

gether

-0.15

POSITIVE LOGITS

orem

0.31

/Sub

0.20

/Set

0.18

.fromCharCode

0.17

noDB

0.17

 WithEvents

0.17

notated

0.17

/Branch

0.16

_FILENO

0.16

semble

0.16

Activations Density 0.428%

code-related structures, particularly arrays, functions, and object manipulation

No Comments

No Known Activations