INDEX

Explanations

timestamps and time indicators

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

ite

-0.17

arry

-0.15

arity

-0.15

adero

-0.14

\Carbon

-0.14

rne

-0.14

apan

-0.14

ipo

-0.14

UDGE

-0.14

ÃŃÅĻ

-0.14

POSITIVE LOGITS

Imm

0.18

Imm

0.17

_imm

0.15

IST

0.15

rok

0.14

linkplain

0.14

-imm

0.14

("(%

0.14

0.13

ropri

0.13

Activations Density 0.014%

timestamps and time indicators

No Comments

No Known Activations