INDEX

Explanations

phrases related to past events or occurrences

oai_token-act-pair · gpt-3.5-turbo

references to previous seasons or years in a context related to performance metrics

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

okers

-0.73

ocks

-0.71

oller

-0.70

bsite

-0.69

alth

-0.68

apse

-0.67

coins

-0.66

ongs

-0.66

ordes

-0.66

IRC

-0.65

POSITIVE LOGITS

 year

1.45

 week

1.41

 month

1.33

 night

1.30

 season

1.30

 weekend

1.23

 semester

1.16

 summer

1.10

 offseason

1.05

 winter

1.04

Activations Density 0.061%

phrases related to past events or occurrences

references to previous seasons or years in a context related to performance metrics

No Comments

No Known Activations