INDEX

Explanations

phrases related to personal experiences and interactions

oai_token-act-pair · gpt-3.5-turbo

narratives and personal stories related to experiences and interactions with others

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 awaits

-0.80

joice

-0.71

 await

-0.69

 warns

-0.69

 Parables

-0.66

 wield

-0.65

 foes

-0.65

 urges

-0.64

 welcomes

-0.63

 recy

-0.63

POSITIVE LOGITS

didn

0.96

 hindsight

0.94

0.92

['

0.91

Initially

0.90

went

0.90

 Came

0.89

Eventually

0.88

was

0.87

 Ended

0.86

Activations Density 0.633%

phrases related to personal experiences and interactions

narratives and personal stories related to experiences and interactions with others

No Comments

No Known Activations