INDEX

Explanations

phrases related to cost or cost-effectiveness

oai_token-act-pair · gpt-3.5-turbo

terms associated with low-cost options and solutions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

atars

-0.74

MET

-0.72

ICLE

-0.71

olphin

-0.71

frey

-0.70

ynthesis

-0.70

MAC

-0.69

wered

-0.67

iph

-0.67

ortmund

-0.66

POSITIVE LOGITS

(<

0.80

yet

0.75

 earners

0.69

nered

0.69

 presses

0.68

 distractions

0.68

 setups

0.68

 offenders

0.67

 replacements

0.66

 itiner

0.66

Activations Density 0.065%

phrases related to cost or cost-effectiveness

terms associated with low-cost options and solutions

No Comments

No Known Activations