INDEX

Explanations

words related to significant numbers or quantities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ABCDEFGHIJKLMNOP

-0.16

nze

-0.15

rien

-0.15

usk

-0.15

ussy

-0.15

APPER

-0.15

unge

-0.14

$MESS

-0.14

ancellable

-0.14

ze

-0.13

POSITIVE LOGITS

--

0.17

—for

0.16

 Demir

0.15

—that

0.15

--

0.15

Mir

0.15

 Merrill

0.15

---

0.14

kb

0.14

alin

0.14

Activations Density 0.033%

words related to significant numbers or quantities

No Comments

No Known Activations