INDEX

Explanations

phrases related to explanations or explorations

oai_token-act-pair · gpt-3.5-turbo

instances of the word "explore" or its variations that indicate an examination or investigation of topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

chal

-0.63

 Amid

-0.63

 faces

-0.61

yards

-0.60

 petitions

-0.59

 stand

-0.59

DAY

-0.58

 thumbs

-0.58

 tune

-0.56

 pledge

-0.56

POSITIVE LOGITS

oit

1.56

orers

1.54

oded

1.40

oding

1.39

osion

1.38

icit

1.36

orer

1.35

oration

1.32

ained

1.31

aining

1.25

Activations Density 0.028%

phrases related to explanations or explorations

instances of the word "explore" or its variations that indicate an examination or investigation of topics

No Comments

No Known Activations