INDEX

Explanations

occurrences of the preposition "in."

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

egr

-0.16

venile

-0.15

Tile

-0.14

 classical

-0.14

 entertain

-0.14

uin

-0.14

/Admin

-0.13

 Tile

-0.13

 vener

-0.13

 historic

-0.13

POSITIVE LOGITS

 Rooney

0.29

ell

0.20

 Sally

0.18

Connell

0.17

 Normal

0.16

SuccessListener

0.16

.Normal

0.16

ellipsis

0.15

allocate

0.15

0.14

Activations Density 0.000%

occurrences of the preposition "in."

No Comments

No Known Activations