INDEX

Explanations

general terms referring to collections of people

oai_token-act-pair · gpt-3.5-turbo

phrases emphasizing inclusivity and the well-being of various groups of people

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Collider

-0.69

 bluff

-0.69

USS

-0.67

OLOG

-0.65

 brisk

-0.64

inventoryQuantity

-0.64

Alias

-0.63

 Haunted

-0.61

 Jump

-0.60

 caution

-0.59

POSITIVE LOGITS

 irrespective

0.84

 alike

0.82

soever

0.80

 harmed

0.77

selves

0.75

dden

0.73

 effected

0.73

igent

0.73

folk

0.73

omever

0.73

Activations Density 0.301%

general terms referring to collections of people

phrases emphasizing inclusivity and the well-being of various groups of people

No Comments

No Known Activations