INDEX

Explanations

phrases describing experiences or actions that are a mix of positive and negative emotions

oai_token-act-pair · gpt-3.5-turbo

phrases that express consistency or permanence in experiences or qualities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 CLSID

-0.77

>>\

-0.63

yet

-0.63

ettlement

-0.61

FINEST

-0.59

 analogous

-0.58

DIT

-0.58

 ratification

-0.58

ALLY

-0.58

·

-0.58

POSITIVE LOGITS

ounters

0.71

 surprises

0.69

 underdog

0.68

oglu

0.67

 whichever

0.65

 lurking

0.63

theless

0.63

 paran

0.62

 pursu

0.61

andro

0.61

Activations Density 0.201%

phrases describing experiences or actions that are a mix of positive and negative emotions

phrases that express consistency or permanence in experiences or qualities

No Comments

No Known Activations