INDEX

Explanations

time indicators or timestamps

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

amt

-0.19

lsen

-0.15

eson

-0.15

bidden

-0.14

bz

-0.14

Hus

-0.14

athan

-0.14

.unbind

-0.14

enate

-0.14

tridge

-0.14

POSITIVE LOGITS

/pm

0.19

ET

0.19

OD

0.15

pst

0.15

PDT

0.15

outu

0.15

tember

0.15

.compiler

0.14

nou

0.14

Activations Density 0.009%

time indicators or timestamps

No Comments

No Known Activations