INDEX

Explanations

patterns in equality and inequality comparisons in code

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

 GetLastError

-0.15

examples

-0.14

alan

-0.14

onus

-0.14

Î¹Î±Î½

-0.14

reh

-0.14

unma

-0.14

chedulers

-0.14

Dot

-0.14

qq

-0.14

POSITIVE LOGITS

ople

0.19

ington

0.15

uddenly

0.15

iban

0.14

esteem

0.14

ãĤ¹ãĤ«

0.14

 nhau

0.14

emas

0.14

 null

0.13

Subset

0.13

Activations Density 0.051%

patterns in equality and inequality comparisons in code

No Comments

No Known Activations